conference abstract - icebe · k0037: using multiple machine learning algorithms for cancer...

80
ICEBE 2020 CONFERENCE ABSTRACT - 1 - CONFERENCE ABSTRACT 2020 6th International Conference on Environment and Bio-Engineering (ICEBE 2020) January 19-22, 2020 Uji Obaku Plaza, Uji Campus, Kyoto University, Kyoto, Japan Organized by Supported by Published and Indexed by http://www.icebe.org/

Upload: others

Post on 16-Jul-2020

6 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 1 -

CONFERENCE ABSTRACT

2020 6th International Conference on Environment and

Bio-Engineering (ICEBE 2020)

January 19-22, 2020

Uji Obaku Plaza, Uji Campus, Kyoto University, Kyoto, Japan

Organized by

Supported by

Published and Indexed by

http://www.icebe.org/

Page 2: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 2 -

Conference Venue

Uji Obaku Plaza, Uji Campus, Kyoto University

Website: http://www.uji.kyoto-u.ac.jp/campus/obaku.html

Address: Gokasho, Uji, Kyoto 611-0011

Tel: 0774-38-3021

Email: [email protected]

1. How to Get Here?

The access/map is available from: https://www.kuicr.kyoto-u.ac.jp/sites/icr/about/access/.

2. Hotel Recommendation

Tips: The registration fee does not cover the accommodation. It should be booked by

participants themselves. The follows are some hotels for recommendation. It is suggested an

early booking should be done.

A. New Miyako Hotel

https://www.miyakohotels.ne.jp/newmiyako/english/index.html

B. Kyoto Century Hotel

https://www.keihanhotels-resorts.co.jp/kyoto-centuryhotel/english/

C. APA Hotel Kyoto-Ekimae

https://www.apahotel.com/hotel/kansai/01_kyoto-ekimae/

D. El Inn Kyoto

https://www.elinn-kyoto.com/en/

E. Uji Dai-ichi Hotel

http://ujidai1.jp/ (Japanese page only)

Page 3: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 3 -

Map

Page 4: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 4 -

Table of Contents Introduction 11

Conference Committee 12

Program-at-a-Glance 14

Presentation Instruction 16

Keynote Speaker Introduction 17

Invited Speaker Introduction 24

Oral Session on January 20, 2020

Session 1: Computational Biology

K0016: LTR-TCP-FR: Protein Fold Recognition based on Learning to Rank

Yulin Zhu and Bin Liu

27

K0069: A Mono-bidomain Electrophysiological Simulation Method for Electrical

Defibrillation Research

Jianfei Wang, Lian Jin, Weiqi Wang and Xiaomei Wu

27

K0042: Using Gene-level to Generalize Transcript-level Classification Performance

on Multiple Colorectal Cancer Microarray Studies

Hendrick Gao-Min Lim and Yuan-Chii Gladys Lee

28

K0048: Efficient GPU Acceleration for Computing Maximal Exact Matches in Long

DNA Reads

Nauman Ahmed, Koen Bertels and Zaid Al-Ars

28

K2034: Spectral Structure and Nonlinear Dynamics Properties of Long-term

Interstitial Fluid Glucose

Junichiro Hayano, Atsushi Yamada, Yutaka Yoshida, Norihiro Ueda and Emi

Yuda

29

K0039: Assessing Information Quality and Distinguishing Feature Subsets for

Molecular Classification

Hung-Yi Lin and Da-Yi Yen

29

Session 2: Pharmaceutical Science

K0014: Docking Simulation of Chemerin-9 and ChemR23 Receptor

Keiichi Nobuoka, Hironao Yamada, Takeshi Miyakawa, Ryota Morikawa, Takuya

Watanabe and Masako Takasu

31

K0021: Drug Discovery for Kidney Chromophobe using Molecular Docking

Aysegul Caliskan and Kazim Yalcin Arga

31

K0036: Docking Small Molecules Belonging Laserpitium sp. into LCK Protein 32

Page 5: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 5 -

Upregulated in Rheumatoid Arthritis

Meltem Gulec, Aysegul Caliskan and Ahmet Cenk Andac

K2033: Antimicrobial Property and Mechanism of Action of Mangiferin from

Curcuma Amada

Sayanti Dey, Abhishek Raj, Hari Om and Debjani Dutta

32

K2021: In situ Transfection through Wound Dressing to Promote Tissue

Regeneration

Wei-Wen Hu and Yu-Ting Lin

33

K0029: Theoretical Prediction of Parallel Artificial Membrane Permeability Assay

Permeability using a Two-QSAR Approach

Max K. Leong

33

Session 3: Biological Science and Information Technology

K1007: Effects of 2, 3', 4, 4', 5-pentachlorobiphenyl Exposure during Pregnancy dn

Germ Cell Development and Epigenetic Modification of F1 Mice

Qi-Long He, Xu-Yu Wei, Qian Zhou, Hai-Quan Wang and Shu-Zhen Liu

35

K2027: Morphological Variation and Systematic Value of Indumentum in Philippine

Vavaea Bentham (Meliaceae) Species

Dhiocel A. Celadiña and Edwino S. Fernando

35

K0043: Deep Learning based System to Extract Agricultural Workers‘ Physical

Timeline Data for Acceleration and Angular Velocity

Shinji Kawakura and Ryosuke Shibasaki

36

K0013: A Method for the Inverse QSAR/QSPR based on Artificial Neural Networks

and Mixed Integer Linear Programming

Rachaya Chiewvanichakorn, Chenxi Wang, Zhe Zhang, Aleksandar Shurbevski,

Hiroshi Nagamochi and Tatsuya Akutsu

36

K2014: An Investigation of Factors Influencing Performance of RFID-based Vehicle

Detection

Li-Fei Chen, Ruei-Shiuan Tsai, Yuan-Jui Chang and Chao-Ton Su

37

K2020: Energy Monitoring and Management System Design based on Embedded

Structure

Horng-Lin Shieh, Cheng-Chien Kuo and Sheng-Bo Gao

37

K0046: GpemDB: A Scalable Database Architecture with the Multi-omics

Entity-relationship Model for Integrating Heterogeneous Big-data in Precise Crop

Breeding

Liang Gong, Shengzhe Fan and Wei Wu

38

Session 4: Environmental Chemistry and Materials

K3002: Methotrexate Degradation by UV-C and UV-C/TiO2 Processes with and

without H2O2 Addition on Pilot Reactors

L. A. González-Burciaga, J. C. García-Prieto, C. M. Núñez-Núñez, M.

García-Roig and J. B. Proal-Nájera

39

Page 6: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 6 -

K2015: Empirical Approach for Modeling of Partition Coefficient on Lead

Concentrations in Riverine Sediment

Saadia Bouragba, Katsuaki Komai and Keisuke Nakayama

39

K2023: Potassium Fluoride-doped CaMgAl Layered Double Hydroxides for HCl

Adsorption in the Presence of CO2 at Medium-high Temperature

Wei Wu, Tongwei Wang, Decheng Wang and Baosheng Jin

40

K2025: Quantum Chemical Research on Mechanism of Hydrogen Chloride Removal

by Calcium Carbonate

Jie Zhu, Wei Wu and Dejian Shen

40

K2024: Bifunctional MOFs as Efficient Catalysts for Catalytic Conversion of

Carbon Dioxide into Cyclic Carbonates and DFT Studies

Yuanfeng Wu and Guomin Xiao

41

K1013: A Novel Deep-blue Thermally Activated Delayed Fluorescence Emitter

Dimethylacridine/Thioxanthene-S, S-Dioxide for Highly-efficient Organic

Light-emitting Devices

Young Pyo Jeon

41

Oral Session on January 21, 2020

Session 5: Bioinformatics

K0077: A Sparse Bayesian Approach to Combinatorial Feature Selection and Its

Applications to Biological Data

Ryoichiro Yafune, Daisuke Sakuma, Yasuo Tabei, Noritaka Saito, Einoshin

Suzuki and Hiroto Saigo

43

K0017: Analyses of Interaction between Platinum Bonded LARFH and Gold Surface

by Molecular Dynamics Simulation

Mao Watabe, Keiichi Nobuoka, Hironao Yamada, Takeshi Miyakawa, Ryota

Morikawa, Masako Takasu, Tatsuya Uchida and Akihiko Yamagishi

43

K0038: Identification of Redox-sensitive Cysteines using Sequence Profile

Information

Md. Mehedi Hasan and Hiroyuki Kurata

44

K0019: Relationship between Dynamics of Structures and Dynamics of Hydrogen

Bonds in Hras-GTP/GDP Complex

Takeshi Miyakawa, Kimikazu Sugimori, Kazutomo Kawaguchi, Masako Takasu,

Hidemi Nagao and Ryota Morikawa

44

K0079: Bayesian Optimization for Sequence Data

Kohei Oyamada and Hiroto Saigo

44

K0020: Microorganism Hydrolysate Reduced Lipo-oxidation in Regulated

High-fructose Induced Nonalcoholic Fatty

Chang-Chi Hsieh and Li-Jia Huang

45

K0045: Improvement of Protein Stability Prediction by Integrated Computational

Approach

45

Page 7: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 7 -

Chi-Wei Chen, Meng-Han Lin, Hsung-Pin Chang and Yen-Wei Chu

Session 6: Biomedicine

K0034: Expression of TLR4 Signaling in Acute Appendicitis and Experimental

Peritonitis

Han-Ping Wu

47

K0049: Mapping the Humoral Immune Response of Patients with American

Trypanosomiasis using Phage Display Technology

André A. R. Teixeira, Luis A. Rodriguez-Carnero, Edécio Cunha-Neto, Walter

Colli and Ricardo J. Giordano

47

K1009: External Harmful Substances Induced Sensitization and Its Association

Assessments of Inflammation Mechanisms in Epithelial Cell and Animal Model

Zhi-Yao Ke, Thung-Shen Lai and En-Chih Liao

48

K2017: The Molecular Mechanism and Drug Therapy of Heart Failure

Yifan Su

49

K2003: Rapid Dysphagia Diagnostic Test based on Voice Features

Hsin-Yu Chou, Li-Ya Lin, Li-Chun Hsieh, Pei-Yi Wang and Shih-Tsang Tang

49

K0059: Corpus Construction of Precision Medicine

Xuejing Ren, Xinying An and Shaoping Fan

49

K0040: Computational Drug-target Interaction Prediction based on Graph

Embedding and Graph Mining

Maha A. Thafar, Somayah Albaradie, Rawan S. Olayan, Haitham Ashoor,

Magbubah Essack and Vladimir B. Bajic

50

Session 7: Medical Informatics

K0012: Rapid Detection and Prediction of Influenza A Subtype using Deep

Convolutional Neural Network based Ensemble Learning

Yu Wang, Junpeng Bao, Jianqiang Du and YongFeng Li

52

K0015: Predicting lncRNA-disease Association based on Extreme Gradient Boosting

Xi Tang, Menglu Li, Wei Zhang and Junfeng Xia

52

K0041: Breast Cancer Subtype by Imbalanced Omics Data through A Deep Learning

Fusion Model

Jingwen Zeng, Hongmin Cai and Tatsuya Akutsu

53

K2006: Design and Implementation of NoSQL-based Medical Image Archive

Underlaying FHIR

Chien-Hua Chu, Shih-Tsang Tang and Chung-Yueh Lien

53

K0066: A Novel Feature Selection and Classification Method of Alzheimer's Disease

based on Multi-features in MRI

Peiqi Luo, Guixia Kang and Xin Xu

54

K0071: Tracking and Analysis of FUCCI-labeled Cells based on Particle Filters and

Time-to-event Analysis

54

Page 8: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 8 -

Kenji Fujimoto, Shigeto Seno, Hironori Shigeta, Tomohiro Mashita, Junichi

Kikuta, Masaru Ishii and Hideo Matsuda

K2004: Home Wound Dressing Photo Management Cloud System

Meng-Jie Su, Yu-Tong Liu, Szu-Hsien Wu, Ming-Liang Hisiao and Shih-Tsang

Tang

55

K0076: Development of Photo-activatable Insulin Receptor for Insulin Signalling

Pathway Control

Bryan John J. Subong, Mizuki Endo and Takeaki Ozawa

55

Session 8: Preventive Medicine and Rehabilitation Medicine

K3003: Variability of Local Weather as Early Warning for Dengue Hemorrhagic

Fever Outbreak in Indonesia

Khaidar Ali, Isa Ma’rufi, Wiranto and Anis Fuad

57

K2035: Associations between Seasonal Variation of Heart Rate Variability and

Healthy Life Expectancy in Japan

Emi Yuda, Masaya Kisohara, Yutaka Yoshida, Norihiro Ueda and Junichiro

Hayano

57

K0063: Exploring Elders‘ Willingness and Needs for Adopting an Interactive

Somatosensory Game into Muscle Rehabilitation Systems

Yu-Ling Chu, Chien-Hsiang Chang, Yi-Wen Wang, Chien-Hsu Chen and

Yang-Cheng Lin

58

K2005: The Smartphone-based Rehabilitation Assistant for Stroke Patient

Pei-Yu Lin, Chieh-Min Lin, Chih-Liang Wu, Kuo-Feng Chou and Shih-Tsang Tang

59

K0060: Dynamically Colour Changing Actuator for Cyanosis Baby Manikin

Application with the Philips Hue LED Kit

Azmi Nur Fatihah, Frank Delbressine and Loe Feijs

59

K2007: A Deep Image based Proposal for Posture Classification

Wan-Chun Liao, Jing Wang, Yi-Shu Zhou, Ying-Hui Lai, Wai-Keung Lee and

Shih-Tsang Tang

60

K0002: Rational Discovery of Human Superoxide Dismutase I (hSOD1)

Modulators: A Potential Therapeutic Target for Amyotrophic Lateral Sclerosis (ALS)

Balasundaram Padmanabhan

60

K0031: Orai3 Contributes to the Migration Capacity in Lobular Liver Cancer

Jing Yan, Xiuliang Zhao, Xia Liu, Zhaodi Gao and Ting Wei

61

Poster Session on January 20, 2020

Poster Session 1: Biochemistry and Chemical Engineering

K0009: Mutation of Microbial Glycosyltransferase for Efficient and Regioselective

Biosynthesis of Rare Ginsenoside Rh1

Lu Zhao, Jianlin Chu and Bingfang He

62

K0061: The Biological Effects of Haematococcus Pluvialis Extracts on Fibroblasts 62

Page 9: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 9 -

Lin Jun-Yan, Liu Wei-Chung adn Keng Nien-Tzu

K0010: An Efficient Synthesis of Nucleoside Analogues Catalyzed by A Novel

Thermostable Purine Nucleoside Phosphorylase from Aneurinibacillus Migulanus

AM007

Gaofei Liu, Tiantong Cheng, Jianlin Chu, Sui Li and Bingfang He

63

K0031: Repeated-batch Lactic Acid Fermentation using a Novel Bacterial

Immobilization Technique based on a Microtube Array Membrane

Chien-Chung Chen, Chuan-Chi Lan and Hong-Ting Victor Lin

63

K0011: A Simple Enhancer for Efficient Extracellular Expression of Human

Epidermal Growth Factor Secreted by E. Coli

Lupeng Cui, Yumeng Qiu and Bingfang He

64

K0065: Retinoic Acid Receptor - and Retinoid X Receptor -specific Agonists are

Hemodynamics-based Therapeutic Components for Vascular Disorders

Ting-Yu Lee

64

K0044: The Mechanical Behaviors of Polyethylene/Silver Nanoparticle Composites:

An Insight from Molecular Dynamics Study

Chuan Chen and Shin-Pon Ju

65

K2010: Characteristics of Hydrocarbons and Carbonyl Compounds Emitted from the

Oil Fume of Restaurant Oil Trap

Chia-Hsiang Lai, Yao-Kai Xiang and Zhen-Hong Zhu

65

K3004: Research on the Coordinated Development of Energy-environment System

Zhang Song

66

K0018: Optimization of Cold-adapted α-amylase Production in Escherichia Coli by

Regulation of Induction Conditions and Supplement with Osmolytes

Xiaofei Wang, Guangfeng Kan, Xiulian Ren, Cuijuan Shi and Ruiqi Wang

66

K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung

Adenocarcinoma

Le Wei, Wanning Wen and Zhou Fang

67

K1004: Identification of the Association between Hepatitis B Virus and Liver

Cancer using Machine Learning Approaches based on Amino Acid

Zhaoyang Cao

67

K0070: Computational Modeling of Myocardial Thermal Lesion Induced by

Multi-source Frequency Control RF Ablation Method

Shengjie Yan, Kaihao Gu, Weiqi Wang and Xiaomei Wu

68

Poster Session on January 21, 2020

Poster Session 2: Systematic Biology

K0083: CitAB Two-component System-regulated Citrate Utilization Contributes to

Vibrio Cholerae Competitiveness with the Gut Microbiota

Ming Liu

70

Page 10: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 10 -

K2002: Preparation and Application of Chitosan-phytic Acid Complex Gel Beads

Shao-Jung Wu, Zhi-Run Chen and Chi-Chen Kuo

70

K0084: The 58th Cysteine of TcpP is Essential for Vibrio Cholerae Virulence Factor

Production and Pathogenesis

Mengting Shi

71

K0053: Laboratory Solution Service using ExTOPE® IoT Portal

Kazuyuki Kato, Zentaro Kadota, Shoichi Uemura, Masataka Okayama, Hiroshi

Yamaguchi and Tomohiro Araki

71

K1012: Procleave: A Bioinformatic Approach for Protease-specific Substrate

Cleavage Site Prediction by Combining Sequence and Structural Information

Fuyi Li, Tatsuya Akutsu, Jian Li and Jiangning Song

72

K1005: BnMTP3, An Intracellular Transporter from Brassica Napus Confer Zn and

Mn Tolerance in Arabidopsis Thaliana

Dongfang Gu, Xueli Zhou, Xi Chen and Wei Zhang

72

K0055: Comparing Dissimilarity Metrics for Clustering Gene into Functional

Modules using Machine Learning

Xin Yan and Dantong Lyu

73

K1006: OsGCT1, A Novel Gene Related to Cadmium, Copper Tolerance and

Accumulation in Rice

Ending Xu, Xi Chen and Wei Zhang

73

K1001: Rectal Methods of Delivery of Medical Drugs of the Protein Nature

J. K. Ukibayev, U.M. Datkhayev, A.P. Frantsev and D.A. Myrzakozha

74

K2018: The Application of Designated Manager System on Protection Forest

Management at Saga Prefecture, Japan - from the Public-private Partnership and

Policy Implementation Approaches

Ching Li, Chiachi Cheng and Yi-Chine Chu

74

K0067: Automatic Brain Mask Segmentation for Mono-modal MRI

Yanwu Yang, Chenfei Ye, Xutao Guo, Chushu Yang and Heather T. Ma

75

K0056: The Application and Comparison of Confocal and SIM Imaging System

Yiran Liu

75

K0082: Colon Cancer Inhibition Capacity Evaluation and Mechanism Study of A

Novel Biopolymer Conjugated Gold Nanoparticle System as Chemo-reagent Carrier

Wei-Hung Hung, Kuen-Chan Lee and Er-Chieh Cho

76

Academic Visit 77

Note 79

Page 11: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 11 -

Introduction

Welcome to 2020 6th International Conference on Environment and Bio-Engineering (ICEBE 2020). It is organized by Biology and Bioinformatics Society (BBS) under Hong Kong Chemical, Biological & Environmental Engineering Society (CBEES) and supported by Kaohsiung Medical University and Kyoto Convention & Visitors Bureau.

ICEBE is an interdisciplinary international conference that invites academics and independent scholars and researchers from around the world to meet and exchange the latest ideas and views in a forum encouraging respectful dialogue. ICEBE serves as an outstanding platform that gathers all the relevant communities and domains together, an international forum for researchers and industry practitioners to exchange the latest fundamental advances in the state of the art and practice of Environment and Bio-Engineering.

Papers will be published in the following journals:

Journal of Environmental Science and Development (IJESD,

ISSN:2010-0264): indexed by Scopus (Since 2019), Chemical Abstracts

Services (CAS), CABI, Ulrich Periodicals Directory, Electronic Journals

Library, Crossref, ProQuest.

Or

International Journal of Bioscience, Biochemistry and Bioinformatics

(IJBBB, ISSN: 2010-3638), and will be included in the Engineering &

Technology Digital Library, and indexed by WorldCat, Google Scholar,

Cross ref, ProQuest.

Conference website and email: http://www.icebe.org/; [email protected]

Page 12: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 12 -

Conference Committee

General Conference Chairs Prof. Chan Jin Park, Incheon National University, South Korea

Prof. Dahms Hans-Uwe, Kaohsiung Medical University, Taiwan

Prof. Harold Yih-Chi Tan, National Taiwan University, Taiwan

Program Co-Chairs Prof. Daniel Gang, University of Louisiana at Lafayette, USA

Prof. Junichiro Hayano, Nagoya City University, Japan

Prof. Jiann-Shing Shieh, Yuan Ze University, Taiwan

Technical Committee Prof. Sudhanshu Sekhar Panda, University of North Georgia, USA

Prof. S. P. Singh, Jharkhand Rai University, Ranchi, India

Prof. Wonsan Choi, Hanbat National University, South Korea

Prof. Won Sik Shin, Kyungpook National University, South Korea

Prof. Saud Bali Al-Shammari, Public Authority for Applied Education, Kuwait

Prof. Mohamed Ksibi, Higher Institute of Biotechnology of Sfax, Tunisia

Prof. Md. Munsur Rahman, Bangladesh University of Engineering and Technology,

Bangladesh

Prof. Jaebok Lee, Kyungsung University, South Korea

Prof. Isabel Margarida Horta Ribeiro Antunes, Polytechnic Institute of Castelo Branco,

Portugal

Prof. I. M. H. R. Antunes, Polytechnic Institute of Castelo Branco, Portugal

Prof. Hwong-wen Ma, National Taiwan University, Taiwan

Prof. Guoan Wang, China Agricultural University, China

Prof. Fi-John Chang, National Taiwan University, Taiwan

Prof. Chung-Te Lee, National Central University, Taiwan

Prof. Ben Hung-Pin Huang, National Taiwan University, Taiwan

Prof. Anda Valdovska, Latvia University of Agriculture, Latvia

Assoc. Prof. Yu-Chuan Yang, National Dong Hwa University, Taiwan

Assoc. Prof. Yi-Tui Chen, National Taipei University of Nursing and Health Sciences,

Taiwan

Assoc. Prof. Wen-Che Hou, National Cheng Kung University, Taiwan

Assoc. Prof. Wei-Hsiang Chen, National Sun Yat-sen University, Taiwan

Assoc. Prof. Peng-Ting Chen, National Cheng Kung University, Taiwan

Assoc. Prof. Naomichi Yamamoto, Seoul National University, South Korea

Assoc. Prof. Maria T. D. Albuquerque, Instituto Politécnico de Castelo Branco, Portugal

Assoc. Prof. Chander Prabha, Patna University, India

Assoc. Prof. Alexandre Marco Da Silva, São Paulo State University, Brazil

Assoc. Prof. Abha Singh, Patna University, India

Dr. Pisitpong Intarapong, King Mongkut’s University of Technology Thonburi, Thailand

Dr. Yong-Seok Seo, Kangwon National University, South Korea

Page 13: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 13 -

Dr. Tarek Mohammad Abdolkader, Umm AlQura University, Saudi Arabia

Dr. Oana A.Cuzman, ICVBC-CNR, Italy

Dr. Fabiana Corami, CNR-IDPA, Italy

Dr. Dott. Luigi Berardi, Politecnico di Bari, Italy

Prof. Giovanni Vallini, University of Verona, Italy

Prof. Suzuki Iwane, University of Tsukuba, Japan

Assist. Prof. Nazlina Haiza Mohd Yasin, Universiti Kebangsaan Malaysia, Malaysia

Assoc. Prof. H.S. El-Sheshtawy, Egyptian Petroleum Research Institute, Egypt

Assoc. Prof. Prasoon Kumar Singh, Indian Institute of Technology (ISM), India

Assoc. Prof. Chun-Yang Yin, Newcastle University in Singapore, Singapore

Prof. Ahmad Zuhairi Abdullah, Universiti Sains Malaysia, Malaysia

Assist. Prof. Narong Touch, Tokyo University of Agriculture, Japan

Page 14: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 14 -

Program-at-a-Glance January 19,

2020 (Sunday) 10:00-17:00 Arrival Registration@Room 3 (1F)

January 20,

2020

(Monday)

09:30-17:00 Arrival Registration@Room 3 (1F)

Morning Conference@Room 4&5 (1F)

09:30-09:35

Opening Remarks

Advisory Conference Chair: Prof. Tomohiro Araki, Tokai

University, Japan

09:35-09:45

Welcome Address

General Conference Chair: Prof. Tatsuya Akutsu, Kyoto

University, Japan

09:45-10:25

Keynote Speech I

Prof. TSUI Kwok-Wing Stephen, The Chinese University of

Hong Kong, Hong Kong

Topic: ―The Genomes and Microbiomes of Dermatophagoides

Farinae and Dermatophagoides Pteronyssinus Reveal a Broad

Spectrum of Dust Mite Allergens‖

10:25-10:50 Coffee Break & Group Photo

10:50-11:30

Keynote Speech II

Prof. Jean-Philippe Vert, Google Brain, France and MINES

ParisTech, France

Topic: ―Learning from Single-cell Genomics Data‖

11:30-12:10

Keynote Speech III

Prof. Kuo-Sheng Cheng, National Cheng Kung University,

Taiwan

Topic: ―An Integrated Analysis System for Cephalometric

Applications‖

12:10-12:30

Invited Speech I

Assoc. Prof. Jiangning Song, Monash University, Australia

Topic: ―Leveraging the Power of Data-Driven Machine Learning

Techniques to Address Significant Biomedical Classification

Problems‖

12:30-13:30 Lunch@Kihada Restaurant (2F)

Afternoon Conference

13:30-15:00

Session 1@Room 1 (1F)

Topic: ―Computational

Biology‖

6 presentations

Session 2@Room 2 (1F)

Topic: ―Pharmaceutical

Science‖

6 presentations

15:00-16:00 Coffee Break & Poster Session

Poster Session 1@Lobby of Room 3 (1F)

Topic: ―Biochemistry and Chemical Engineering‖ 13 presentations

Page 15: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 15 -

16:00-17:45

Session 3@Room 1 (1F)

Topic: ―Biological Science and

Information Technology‖

7 presentations

Session 4@Room 2 (1F)

Topic: ―Environmental

Chemistry and Materials‖

7 presentations

January 21,

2020

(Tuesday)

09:30-17:00 Arrival Registration@Room 3 (1F)

Morning Conference@Room 4&5 (1F)

09:30-09:35

Opening Remarks

Advisory Conference Chair: Prof. Tomohiro Araki, Tokai

University, Japan

09:35-10:15

Keynote Speech IV

Prof. Hans-Uwe Dahms, Kaohsiung Medical University, Taiwan

Topic: ―Which Toxicity Evaluations Provide Better Risk

Assessments–in situ, in vivo, in vitro, or in silico?‖

10:15-10:40 Coffee Break & Group Photo

10:40-11:20

Keynote Speech V

Prof. Jiann-Shing Shieh, Yuan Ze University, Taiwan

Topic: ―Intelligent Signal Processing in Autonomous Systems for

Healthcare Monitoring and Control‖

11:20-11:40

Invited Speech II

Prof. Sung Wing Kin, Ken, National University of Singapore,

Singapore

Topic: ―Improving CNV Calling from High-Throughput

Sequencing Data through Statistical Testing‖

11:40-12:00

Invited Speech III

Prof. Ryoji Nagai, Tokai University, Japan

Topic: ―Protein Modification by Metabolic Intermediates and its

Involvement with Life-Style Related Diseases‖

12:00-13:30 Lunch@Kihada Restaurant (2F)

Afternoon Conference

13:30-15:15

Session 5@Room 1 (1F)

Topic: ―Bioinformatics‖

7 presentations

Session 6@Room 2 (1F)

Topic: ―Biomedicine‖

7 presentations

15:15-16:15 Coffee Break & Poster Session

Poster Session 2@Lobby of Room 3 (1F)

Topic: ―Systematic Biology‖ 13 presentations

16:15-18:15

Session 7@Room 1 (1F)

Topic: ―Medical Informatics‖

8 presentations

Session 8@Room 2 (1F)

Topic: ―Preventive Medicine

and Rehabilitation Medicine‖

8 presentations

18:30-20:30 Dinner@Kihada Restaurant (2F)

January 22,

2020

(Wednesday)

09:30-17:30 Academic Visit

Tips: Please arrive at the Conference Room 10 minutes before the session begins to upload PPT into

the laptop; submit the poster to the staff when signing in.

Page 16: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 16 -

Presentation Instruction

Instruction for Oral Presentation

Devices Provided by the Conference Organizer:

Laptop Computer (MS Windows Operating System with MS PowerPoint and Adobe Acrobat

Reader); Digital Projectors and Screen; Laser Stick

Materials Provided by the Presenters:

PowerPoint or PDF Files (Files should be copied to the Conference laptop at the beginning of

each Session.)

Duration of each Presentation (Tentatively):

Keynote Speech: about 35 Minutes of Presentation and 5 Minutes of Question and Answer

Invited Speech: about 15 Minutes of Presentation and 5 Minutes of Question and Answer

Oral Presentation: about 12 Minutes of Presentation and 3 Minutes of Question and Answer

Poster Presentation: about 3 Minutes of Presentation and 2 Minutes of Question and Answer

Instruction for Poster Presentation

Materials Provided by the Conference Organizer:

The place to put poster

Materials Provided by the Presenters:

Home-Made Posters: Submit the poster to the staff when signing in; Poster Size: A1

(841*594mm); Load Capacity: Holds up to 0.5 kg

Best Presentation Award One Best Oral or Poster Presentation will be selected from each session, and the Certificate

for Best Presentation will be awarded at the end of the session on January 20-21, 2020.

Dress Code Please wear formal clothes or national representative of clothing.

Disclaimer Along with your registration, you will receive your name badge, which must be worn when

attending all conference sessions and activities. Participants without a badge will not be

allowed to enter the conference venue. Please do not lend your name badge to the persons

who are not involved in the conference and do not bring the irrelevant persons into the

conference venue.

The conference organizers cannot accept liability for personal injuries, or for loss or damage

of property belonging to conference participants, either during, or as a result of the conference.

Please check the validity of your own insurance.

Page 17: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 17 -

Keynote Speaker Introduction

Keynote Speaker I

Prof. TSUI Kwok-Wing Stephen

The Chinese University of Hong Kong, Hong Kong

TSUI Kwok-Wing Stephen is currently a Professor in the School of Biomedical Sciences,

the Head of Division of Genomics and Bioinformatics and the Director of Hong Kong

Bioinformatics Centre in the Chinese University of Hong Kong (CUHK). In 1995, he

received his PhD degree in Biochemistry at CUHK. He was then appointed as an Assistant

Professor in the Biochemistry Department in 1997 and promoted to the professorship in 2004.

He was also a former member of the International HapMap Consortium and worked on the

single nucleotide polymorphisms of human chromosome 3p. During the SARS outbreak in

2003, his team was one of the earliest teams that cracked the complete genome of the

SARS-coronavirus. Totally, he has published more than 220 scientific papers in international

journals, including Nature, NEJM, Lancet, PNAS, Circulation, JACI and Genome Biology.

His major research interests are next generation sequencing, bioinformatics, human genetic

diseases and molecular microbiology.

Topic: ―The Genomes and Microbiomes of Dermatophagoides Farinae and

Dermatophagoides Pteronyssinus Reveal a Broad Spectrum of Dust Mite Allergens‖

Abstract—It is well known that house dust mites (HDMs) are predominant sources of inhalant

allergens associated with allergic disease. Therefore, sequenced house dust mite (HDM)

genomes would certainly advance our understanding of HDM allergens, a common cause of

human allergies. To produce annotated Dermatophagoides (D.) farinae and D. pteronyssinus

genomes, we developed a combined genomic-transcriptomic-proteomic approach for the

elucidation of HDM allergens. High quality D. farinae and D. pteronyssinus genomes and

transcriptomes were assembled with high-throughput DNA sequencing platforms including

PacBio, Illumina HiSeq and ion torrent. The mite‘s microbiome composition was at the same

time determined and the predominant genus was validated immunohistochemically. Putative

allergens were then evaluated with immunoblotting, immunosorbent assays, and skin prick

tests. In this study, 79.79-Mb and 66.85-Mb genomes of D. farinae and D. pteronyssinus,

respectively, was constructed. Moreover, the full gene structures of canonical allergens and

non-canonical allergen homologues were produced. Using mass spectrometry analysis of D.

farinae protein spots reactive to pooled sera from HDM-allergic patients, novel major

allergens were found. In D. farinae, the predominant bacterial genus among 100 identified

Page 18: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 18 -

species was Enterobacter (63.4%), among them Enterobacter cloacae and Enterobacter

hormaechei were most predominant. KEGG pathway analysis revealed a phototransduction

pathway in D. farinae as well as thiamine and amino acid synthesis pathways suggestive of an

endosymbiotic relationship between D. farinae and its microbiome. In summary, high quality

HDM genomes produced from genomic, transcriptomic, and proteomic experiments revealed

allergen genes and a diverse endosymbiotic microbiome, providing a tool for further

identification and characterization of HDM allergens and development of diagnostics and

immunotherapeutic vaccines.

Page 19: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 19 -

Keynote Speaker II

Prof. Jean-Philippe Vert

Google Brain, France and MINES ParisTech, France

Jean-Philippe Vert is a research scientist at Google Brain, and adjunct researcher at MINES

ParisTech. After a PhD in mathematics at ENS Paris in 2001 and a post-doc at Kyoto

University, he held various academic positions at MINES ParisTech, Institut Curie, UC

Berkeley and ENS Paris. His main research contributions are in machine learning and

computational biology, in particular in cancer genomics and precisions medicine.

Topic: ―Learning from Single-cell Genomics Data‖

Abstract—Single-cell genomics allows capture the diversity of individual cells at the

molecular level, and has revolutionized our understanding of development processes or tumor

heterogeneity. It also raises numerous modeling and computational challenges. In this talk I

will present some approaches we developed for data normalization, gene network inference

and integration of heterogeneous views from single-cell genomics data.

Page 20: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 20 -

Keynote Speaker III

Prof. Kuo-Sheng Cheng

National Cheng Kung University, Taiwan

Prof. Kuo-Sheng Cheng received his B.Sc, M.Sc, and Ph.D degrees from Department of

Electrical Engineering, National Cheng Kung University, Tainan, TAIWAN. He also received

his M.Sc degree from Department of Biomedical Engineering, Rensselaer Polytechnic

Institute, USA. Currently, he is a professor with the Department of Biomedical Engineering,

National Cheng Kung University. He also is the Director of Department of Maintenance and

Engineering, National Cheng Kung University Hospital and the Director of Engineering and

Technology Promotion Center, which is financial supported by Ministry of Science and

Technology, TAIWAN. He was the past President of the Biomedical Engineering Society of

TAIWAN. His research interests includes medical image processing, electrical impedance

imaging and biomedical instrumentation.

Topic: ―An Integrated Analysis System for Cephalometric Applications‖

Abstract—With the rapid advance of information and communication technologies, to develop

the digital as well as smart dentistry becomes an important issue in oral medicine. In the

procedures of conventional cephalometry, many steps rely on the manual processing such as

the landmarking and superimposition. The automation of landmarking and superimposition

are the first step in cephalometric analysis. Those points in cephalograms representing the

anatomical structures of the skull are called landmarks, which are routinely analyzed for

diagnosis and treatment planning. In this integrated analysis system, the image processing

module was developed for locating the landmarks of X-ray cephalogram automatically. The

image was divided into eight rectangular subimages that containing all the useful landmarks.

A genetic algorithm combined with perceptron was proposed for feature subimage extraction.

All the sugimages were enhanced in the preprocessing stage. The pyramid method was

applied to reduce the resolution of image, and the edges were detected by the appropriate edge

detectors or the best orientation edge detector. The curve of each edge was adjusted elastically

with the pre-stored models. Positions of landmarks could be then located immediately and the

associated parameters could also be computed for diagnosis. Secondly, the analysis of the

spatial changes of the craniofacial structures for orthodontic treatment or surgery always

relies on the superimposition of pre- and post-treatment cephalometric tracings. A

computerized superimposition module was also developed for solving this problem. The

feature curves were detected and traced for the cranial base using the best oriental edge

detector and Hough transform, and for the mandibular using the Laplacian of Gaussian and

Page 21: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 21 -

grouping methods. The superimposition was automated following the clinically available

procedures. From the experimental results, it was shown that the cephalometric analysis may

be improved with its accuracy and processing time using this proposed integrated analysis

syste.

Page 22: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 22 -

Keynote Speaker IV

Prof. Hans-Uwe Dahms

Kaohsiung Medical University, Taiwan

Dr. Hans-Uwe Dahms is a professor at Kaohsiung Medical University. He is interested in

stress responses in general and within aquatic systems in particular. He, his colleagues and

students integratively study pollution and the toxicology of stressors from physical, chemical,

and biological sources. He is equally interested in climate change, the spread of diseases,

antibiotic-resistance, food and drink safety from water sources, and integra-tive approaches in

environmental and public health monitoring, risk assessment and management. He advised

more than 25 Ph.D. students in their research and published more than 275 papers in scientific

journals. He served as a reviewer for more than 70 SCI journals, as editorial board member of

12 reputed scientific journals, academic editor of PLosONE, and as editor in chief of

FRONTIERS in Marine Pollution.

Topic: ―Which Toxicity Evaluations Provide Better Risk Assessments–in situ, in vivo, in

vitro, or in silico?‖

Abstract—Describing the toxicological profile of a substance is the first step required for risk

assessments. Among a wide range of approaches are in vitro methods widely used to

characterise toxicological properties including toxicokinetics with regulatory acceptance

mainly confined to in vitro tests which investigate genotoxic endpoints. Chemoinformatics

refers to in silico approaches that make use of computer applications or computer simulations.

In silico predictive models generally provide fast and economic screening tools for compound

properties. They allow a high throughput and a constant optimization. They are less expensive,

less time consuming, have a high reproducibility, and reduce experimental efforts.

Computational approaches can also prioritize chemicals for their toxicological evaluation in

order to reduce the amount of costly in vitro and ethically problematic in vivo toxicological

screenings, and provide early alerts for newly developed substances. Limitations include that

ADME aspects (absorption, distribution, metabolism, and excretion – which are basic

pharmacokinetic descriptors) are difficult to consider. The programs, descriptors, and

applicabilities are sometimes not clear. In addition are carcinogenicity predictions only

possible when genotoxic compounds are considered. An approach is developed here that

selects appropriate methods in a multiple weight of evidence.

Page 23: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 23 -

Keynote Speaker V

Prof. Jiann-Shing Shieh

Yuan Ze University, Taiwan

Jiann-Shing Shieh received the B.S. and M.S. degrees in chemical engineering from

National Cheng Kung University, Taiwan, in 1983 and 1986, respectively, and the Ph.D

degree in automatic control and systems engineering from The University of Sheffield, U.K.

in 1995. He is currently a Professor with the Department of Mechanical Engineering, also a

Joint Professor with the Graduate School of Biotechnology and Bioengineering, and also

serves as the Dean of the College of Engineering, Yuan Ze University, Taiwan. His research

interests are focused on biomedical engineering, particularly in bio-signal processing,

intelligent analysis and control, medical automation, pain model and control, critical care

medicine monitoring and control, dynamic cerebral autoregulation research, and brain death

index research.

Topic: ―Intelligent Signal Processing in Autonomous Systems for Healthcare Monitoring

and Control‖

Abstract—Recently, significant developments have been achieved in the field of artificial

intelligence (AI), in particularly the introduction of deep learning technology that has

improved the learning and prediction accuracy to unpresented levels, especially when dealing

with big data and high-resolution images. Significant developments have occurred in the area

of medical signal processing and healthcare monitoring such as vital biological signs for

biomedical systems which are carried out by instruments that generate large data sets, in

addition to the growth in population that has resulted in big data sets that require artificial

intelligence techniques to analyse and model. Hence, we propose an autonomous system

(including not only monitoring sensors from patients, but also modelling, critic, fault

detection, and specification algorithms for patients) which will be the key driving factor of

future healthcare concepts, and together with internet of things (IoT), AI, big data analysis,

edge computing, fog computing, and cloud computing. Definitely, it will open a new era of

intelligent signal processing in autonomous systems for healthcare monitoring and control.

Page 24: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 24 -

Invited Speaker Introduction

Invited Speaker I

Assoc. Prof. Jiangning Song

Monash University, Australia

Dr. Song is an Associate Professor and Group Leader in the Cancer and Infection and

Immunity Programs in School of Biomedical Sciences, Faculty of Medicine, Nursing and

Health Sciences, Monash University, Melbourne, Australia. Trained as a bioinformatician and

data-savvy scientist, he has a very strong specialty in Artificial Intelligence, Bioinformatics,

Comparative Genomics, Cancer Genomics, Computational Biomedicine, Data Mining,

Infection and Immunity, Machine Learning, Proteomics, and 'Biomedical Big Data', which are

highly sought-after expertise and skill sets in the data-driven biomedical sciences. He was

awarded a four-year NHMRC Peter Doherty Biomedical Fellowship. He also received both

the JSPS Long-term and Short-term Fellowships and did his postdoctoral research at the

Bioinformatics Center, Kyoto University, Japan. He is a member of the Monash Centre for

Data Science and also Associate Investigator of the ARC Centre of Excellence in Advanced

Molecular Imaging at Monash University. He is an Associate Editor of BMC Bioinformatics

and Protein & Peptide Letters and serves as an Advisory Board member of Current Protein &

Peptide Science.

Topic: ―Leveraging the Power of Data-driven Machine Learning Techniques to Address

Significant Biomedical Classification Problems‖

Abstract—Recent advances in high-throughput sequencing have significantly contributed to

an ever-increasing gap between the number of gene products (‗proteins‘) whose function is

well characterized and those for which there is no functional annotation at all. Experimental

techniques to determine the protein function are often expensive and time-consuming.

Recently, machine-learning (ML) techniques based on statistical learning have provided

efficient solutions to challenging problems of sequence classification or functional annotation

that were previously considered difficult to address. In this talk, by combining our recent

research progress, I will highlight some important developments in the prediction of two

representative sequence labeling problems in computational biology based on the

high-dimensional, noisy and redundant information derived from sequences and the 3D

structure. I will illustrate how ML methods can extract the predictive power from a variety of

features that are derived from different aspects of the data and useful strategies that help to

contribute to the predictive performance of ML approaches.

Page 25: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 25 -

Invited Speaker II

Prof. Sung Wing Kin, Ken

National University of Singapore, Singapore

Prof. Dr. Wing-Kin Sung received both the B.Sc. and the Ph.D. degree in the Department of

Computer Science from the University of Hong Kong in 1993, 1998, respectively. He is a

professor in the Department of Computer Science, School of Computing, NUS. Also, he is a

senior group leader in Genome Institute of Singapore. He has over 20 years experience in

Algorithm and Bioinformatics research. He also teaches courses on bioinformatics for both

undergraduate and postgraduate. He was conferred the 2003 FIT paper award (Japan), the

2006 National Science Award (Singapore), and the 2008 Young Researcher Award (NUS) for

his research contribution in algorithm and bioinformatics.

Topic: ―Improving CNV Calling from High-throughput Sequencing Data through

Statistical Testing‖

Abstract—Structural variations (SV) are large scale mutations in a genome; although less

frequent than point mutations, due to their large size they are responsible for more heritable

differences between individuals. Two prominent classes of SVs are deletions and tandem

duplications. They play important roles in many devastating genetic diseases, such as

Smith-Magenis syndrome, Potocki-Lupski syndrome and Williams-Beuren syndrome. Since

paired-end whole genome sequencing data has become widespread and affordable, reliably

calling deletions and tandem duplications has been a major target in bioinformatics;

unfortunately, the problem is far from being solved, since existing solutions often offer poor

results when applied to real data. In this talk, we will focuses on detecting deletions and

tandem duplications from paired next-generation sequencing data. We will discuss why

deletions and tandem duplications are difficult to call. Then, we will propose a statistical

method SurVIndel that outperforms existing methods on both simulated and real biological

datasets

Page 26: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 26 -

Invited Speaker III

Prof. Ryoji Nagai

Tokai University, Japan

Dr. Ryoji Nagai received his Bachelor of Science in Teikyo University Science &

Engineering, Master degree in University of Shizuoka and Ph. D degree in Kumamoto

University. He is a Professor in the Laboratory of Food and Regulation Biology, Graduate

School of Agriculture, Tokai University. His research fields are investigation of pathways of

advanced glycation end-products (AGEs) and its inhibitors to prevent lifestyle-related

diseases.

Topic: ―Goodness of Fit Test at Extreme of Disease Risk Distribution‖

Abstract—The number of diabetic patients in the world is continuously increasing. A

measurement of Hemoglobin A1c (HbA1c), which is formed by the reaction between

hemoglobin and blood sugar through the Maillard reaction, is generally adopted as a clinical

marker of blood glucose control in patients with diabetes. In the end, advanced glycation end

products (AGEs) are formed through oxidation, dehydration and condensation. This reaction

progresses not only in processing foods but also in physiological conditions. Detection of

AGEs in samples is still difficult and many studies overestimate AGEs levels in physiological

samples. Furthermore, many pathways have been reported to be involved in the generation of

AGEs in vivo, it is difficult to clarify the relationship between pathology and glycation by

measuring only a single AGE structure. Therefore, monitoring for multiple AGEs in

biological samples by instrumental analyses, such as liquid chromatography tandem mass

spectrometry (LC-MS/MS), is widely done to assess the biological significance of AGEs.

Therefore, in this talk, we would like to explain the advantages of the multiple analysis of

AGEs to demonstrate the protein modification in vivo and its involvement with

lifestyle-related diseases

Page 27: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 27 -

Session 1

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 20, 2020 (Monday)

Time: 13:30-15:00

Venue: Room 1 (1F)

Topic: “Computational Biology”

Session Chair: Prof. Sung Wing Kin, Ken

K0016

Session 1

Presentation 1

(13:30-13:45)

LTR-TCP-FR: Protein Fold Recognition based on Learning to Rank

Yulin Zhu and Bin Liu

Harbin Institute of Technology, Shenzhen, China

Abstract—Fold type reflects the topological patterns of protein core

structures, and fold recognition is one of the most important tasks in the

field of protein structure and function prediction. Previous studies have

shown that Learning to Rank algorithm combined with different features

extracted by different methods is an effective strategy for protein

classification and detection, and protein similarity network can further

improve the performance of the algorithm. In this study, we added the

profile-based features, alignment score features and protein structure

features into the Learning to Rank model, and then constructed a protein

similarity network to transform fold recognition from low similarity

problem to high similarity problem. Finally, a predictor called LTR-TCP-FR

was proposed. The rigorous test on LE benchmark dataset shows that the

LTR-TCP-FR predictor achieved an accuracy of 73.2%. It outperformed all

the other existing computational predictors. Therefore, LTR-TCP-FR is an

effective method for protein fold recognition, and it will become a useful

tool for protein sequence analysis.

K0069

Session 1

Presentation 2

(13:45-14:00)

A Mono-bidomain Electrophysiological Simulation Method for Electrical

Defibrillation Research

Jianfei Wang, Lian Jin, Weiqi Wang and Xiaomei Wu

Fudan University, China

Abstract—Computational simulation is highly useful to study the

mechanisms and methods of electrical defibrillation. However, current

simulation methods have considerable limitations, such as inability to obtain

the virtual electrode polarization and tremendous computational load. In

order to solve these problems, we proposed a Mono-Bidomain simulation

method that synthesizes the characteristics of the passive bidomain model

and the active monodomain model to simulate the effect of active bidomain

Page 28: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 28 -

simulation. The distribution of virtual electrodes is obtained on the passive

bidomain model by applying an electrical shock on the defibrillation

electrodes at the moment of defibrillation, and the transmembrane potential

of every myocardial node is converted into a transmembrane stimulus

current; then the current is applied to the active monodomain model to get

the propagation of ventricular electrical activity after defibrillation. Using

this method, we simulated the changes of electrical activity of point and

plate stimuli schemes on the ideal 2D and 3D tissue models, respectively. It

shows that the simulation result of this method is very close to that of the

active bidomain simulation, and the maximum average error of upstroke

time is less than 2.5 ms, which verifies the effectiveness of the proposed

method. This method can improve the simulation efficiency by greatly

reducing the computational time and flexibly adjusting the electrode

configurations, and provides a new means for the simulation study of

arrhythmia electrotherapy.

K0042

Session 1

Presentation 3

(14:00-14:15)

Using Gene-level to Generalize Transcript-level Classification Performance

on Multiple Colorectal Cancer Microarray Studies

Hendrick Gao-Min Lim and Yuan-Chii Gladys Lee

Taipei Medical University, Taiwan

Abstract—Several classification algorithms have been applied into

microarray studies for colorectal cancer identification. Algorithms such as

naïve bayes, random forest, logistic regression, support vector machine, and

deep learning have been successfully used in previous studies. The accuracy

of these algorithms shown promising result through n-fold validation.

However, most of studies are limited to transcript-level that will implicate to

biased interpretation of classification result due to different microarray

platform entanglement. Therefore, we applied gene-level classification to

generalize transcript-level classification result on multiple colorectal cancer

microarray studies through different classification algorithms including:

naïve Bayes, random forest, logistic regression, support vector machine, and

deep learning. We evaluated classification performance using several

parameters including: accuracy, area under ROC curve, recall and precision.

As the result, we found biased classification result in transcript-level from

multiple microarray studies can be solved through gene-level classification

by applying annotation and merging. In addition, applying batch effect

removal method can make gene-level classification performance slightly

improved. Furthermore, annotation and merging also can be used to solve

another biased result of feature selection in transcript-level.

K0048

Session 1

Presentation 4

(14:15-14:30)

Efficient GPU Acceleration for Computing Maximal Exact Matches in Long

DNA Reads

Nauman Ahmed, Koen Bertels and Zaid Al-Ars

Delft University of Technology, Netherlands

Abstract—The seeding heuristic is widely used in many DNA analysis

Page 29: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 29 -

applications to speed up the analysis time. In many applications, seeding

takes a substantial amount of the total execution time. In this paper, we

present an efficient GPU implementation for computing maximal exact

matching (MEM) seeds in long DNA reads. We applied various

optimizations to reduce the number of GPU global memory accesses and to

avoid redundant computation. Our implementation also extracts maximum

parallelism from the MEM computation tasks. We tested our

implementation using data from the state-ofthe-art third generation Pacbio

DNA sequencers, which produces DNA reads that are tens of kilobases

long. Our implementation is up to 9x faster for computing MEM seeds as

compared to the fastest CPU implementation running on a server-grade

machine with 24 threads. Computing suffix array intervals (first part of

MEM computation) is up to 3x faster whereas calculating the location of the

match (second part) is up to 9x faster. The implementation is publicly

available at https://github.com/nahmedraja/GPUseed.

K2034

Session 1

Presentation 5

(14:30-14:45)

Spectral Structure and Nonlinear Dynamics Properties of Long-term

Interstitial Fluid Glucose

Junichiro Hayano, Atsushi Yamada, Yutaka Yoshida, Norihiro Ueda and

Emi Yuda

Nagoya City University, Japan

Abstract—The spread of continuous indwelling sensors in the subcutaneous

tissue has enabled continuous monitoring of interstitial fluid glucose

concentration (ISFG) under daily activities. This technology is considered to

enable the development of a method for evaluating glycemic control

function by analyzing not only the detailed state of diabetes and other

pathological conditions but also the characteristics of glycemic dynamics.

To clarify the basic fluctuation characteristics of long-term ISFG, the

spectral structure and nonlinear dynamics properties of ISFG obtained by

continuous monitoring for 11 days were analyzed in healthy and diabetic

subjects.

K0039

Session 1

Presentation 6

(14:45-15:00)

Assessing Information Quality and Distinguishing Feature Subsets for

Molecular Classification

Hung-Yi Lin and Da-Yi Yen

National Taichung University of Science and Technology, Taiwan

Abstract—A feature reduction scheme is developed in this study. The

proposed scheme is a hybrid of unsupervised and supervised methods. The

unsupervised process is designed to exclude the irrelevant and useless

information before executing feature selection. Fuzzy clustering algorithm

will categorize all features into different groups. Then, the supervised

process using the target class information will purify the relevant and

informative factors from the approved feature clusters. This purifying

procedure is particularly suitable for high dimensional datasets with

erroneous information. This proposed strategy is to retain the advantage of

Page 30: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 30 -

simplicity and lower selecting time consumption in filter methods while the

disadvantage of high computational cost due to the exponential quantity of

feature subsets in wrapper methods is completely avoided.

Page 31: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 31 -

Session 2

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 20, 2020 (Monday)

Time: 13:30-15:00

Venue: Room 2 (1F)

Topic: “Pharmaceutical Science”

Session Chair: Prof. Max Leong

K0014

Session 2

Presentation 1

(13:30-13:45)

Docking Simulation of Chemerin-9 and ChemR23 Receptor

Keiichi Nobuoka, Hironao Yamada, Takeshi Miyakawa, Ryota Morikawa,

Takuya Watanabe and Masako Takasu

Tokyo University of Pharmacy and Life Sciences, Japan

Abstract—Chemerin-9 is a nonapeptide that corresponds to the

YFPGQFAFS sequence on the C-terminus of Chemerin protein. Recent

clinical and animal studies using mice, it has been recently reported that

Chemerin-9 binds to the ChemR23 receptor and can suppress the

inflammation-related diseases such as arteriosclerosis.

In this study, molecular dynamics simulations were performed with the

Chemerin-9 peptide to identify structures with high and low free energy.

Docking simulations of these structures of Chemerin-9 and ChemR23

receptor were performed, and the docking model with the lowest free energy

and binding energy of Chemerin was identified. For this model, the binding

sites and binding forces were evaluated. Based on these findings, we aim to

provide insights into the development of new drugs that suppress

arteriosclerosis.

K0021

Session 2

Presentation 2

(13:45-14:00)

Drug Discovery for Kidney Chromophobe using Molecular Docking

Aysegul Caliskan and Kazim Yalcin Arga

Istinye University / Marmara University, Turkey

Abstract—Chromophobe renal cell carcinoma (KICH) is one of renal cell

carcinoma types and effective drugs are not available for treatment of this

disease. With the development of technology, computer-based drug

development studies have started to gain importance in order to shed light

on experimental studies. The aim of this study is to discover new drug

candidates for treatment of KICH by using computer-based drug

repositioning and docking analysis techniques. For this purposes, DPP4, one

of the hub and over-expressed proteins of KICH, was chosen for drug

repositioning by taking account its position and function on cell. Crystal

Page 32: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 32 -

structure of DPP4 was fetched from Protein Data Bank. Zinc15 database

was used to determine appropriate drugs which have the potential to bind

DPP4 protein. 12 molecules that have best binding affinity to DPP4 were

determined among 3786 substances taken from Zinc15 database after

molecular docking studies. These molecules are open to experimental

studies for treatment of KICH. Besides Saxagliptin and Dutogliptin are

already commercially available drugs that are used for treatment of type 2

diabetes by selectively inhibiting DPP4. We recommend that these two

drugs should be investigated also for kidney chromophobe.

K0036

Session 2

Presentation 3

(14:15-14:30)

Docking Small Molecules Belonging Laserpitium sp. into LCK Protein

Upregulated in Rheumatoid Arthritis

Meltem Gulec, Aysegul Caliskan and Ahmet Cenk Andac

Istinye University / Istanbul University, Turkey

Abstract—Docking is a increasingly frequent method and developing

technology in pharmaceutical studies assuring the researchers the chance to

dock small-molecule libraries to a macromolecule in order to find proper

compounds with certain biological functions. At the same time a great

number of pharmacognosy studies has given rise to a database in which

active ingredients isolated from a variety of plants are available. Although

some studies have been carried out to determine biological activities of the

active ingredients, there is still much to discover biological activities for the

rest of the molecules in the database. In-vitro determination of biological

activites for each of the molecules is an exhaustive process and the required

expenses are very high. In-silico docking experiments are very helpful in

studying binding interactions between receptors and ligands, reducing the

exhaustive-time required for in-vitro activity studies. In traditional

medicine, some Laserpitium species have been used in the treatment of

rheumatoid arthritis. The main objective of this study is to determine

binding affinities of the active ingredients of the plant genus Laserpitium

toward LCK proto-oncogene, a Src Family Tyrosine Kinase that is

upregulated in rheumatoid artirititis. The active small molecules of

Laserpitum, obtained from the online database Sci-Finder, docked into the

binding site of LCK protein and yielded high binding affinities with some

small molecules. This is a unique study due to resulting a computer-based

docking data for the isolated active compounds obtained by plant isolation

studies which has not conducted. However, the small molecues with high

binding affinites should be further tested in-vitro.

K2033

Session 2

Presentation 4

(14:30-14:45)

Antimicrobial Property and Mechanism of Action of Mangiferin from

Curcuma Amada

Sayanti Dey, Abhishek Raj, Hari Om and Debjani Dutta

National Institute of Technology Durgapur, India

Abstract—Food spoilage is a process where a food product becomes

unsuitable to ingest by the consumer. Globally, spoilage caused by food

Page 33: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 33 -

borne microorganisms still widely affects all types of food and causes food

waste and loss, even in developed countries. Bacteria, yeast and molds are

the common types of microbes that are responsible for the spoilage of a

considerable amount of food products. Traditionally, the crude extracts of

different parts of medical plants including root, stem, flowers, leaves are

widely used for the treatment of many diseases. These plants contain

bioactive compounds such as flavonoids, poly phenol, alkaloids that possess

antimicrobial, antioxidant and antimutagenic activity. Mangiferin is one

such bioactive compound having putative chemo preventive and antioxidant

properties and has the potential of being used as nutraceutical. It is a

poly-phenol compound derived from Anacardiaceae and Gentianaceae

families and primarily found in Mangifera indica. In this study we have used

Curcuma amada (Mango ginger) to extract mangiferin and found the

concentration to be as high as 206 micrograms/ml. Various antimicrobial

assays have been done using crude Curcuma amada against gram (-) ve and

gram (+) ve bacteria like Escherichia coli, Bacillus subtilis and

Staphylococcus aureus. It has been observed that mangiferin has inhibited

the growth of gram (-) ve bacteria and has selective inhibitory effect on

gram (+) ve bacteria. It can be concluded that mangiferin from Curcuma

amada is a potential nurtaceutical and can be used as antimicrobial,

antioxidative agent.

K2021

Session 2

Presentation 5

(14:45-15:00)

In situ Transfection through Wound Dressing to Promote Tissue

Regeneration

Wei-Wen Hu and Yu-Ting Lin

National Central University, Taiwan

Abstract—Chronic wounds may retard the healing process to cause many

risks. Therefore, it is essential to develop a multifunctional wound dressing

to promote tissue regeneration. To fabricate a versatile composite

nanofibrous matrix, sodium alginate and poly (ε-caprolactone) (PCL) were

coelectrospun as composite nanofibers using a dual jet system. Hydrophilic

alginate fibers may provide a moist environment in wound sites. In addition,

PCL were applied to increase mechanical strength and cell adhesion. Silver

nanoparticles were embedded in PCL fibers for long-term release to inhibit

the growth of microorganism. Plasmid DNA encoding platelet-derived

growth factor B (PDGFB) was delivered from composite fibers because this

growth factor is a chemoattractant for neutrophils and can induce the

proliferation and differentiation of fibroblasts. These PDGFB plasmids were

complexed with polyethylenimine (PEI) to form cationic nanoparticles

which may thus be adsorbed onto anionic alginate fibers through

electrostatic interaction. As wound cells adhered to composite fibers, they

can be in situtransfected to continuously express PDGFB. Moreover,

calcium ions in alginate fibers were released to wound sites through ion

exchange to accelerate hemostasis. This comprehensive dressing provides an

ideal solution to heal chronic wounds.

Page 34: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 34 -

K0029

Session 2

Presentation 6

(14:00-14:15)

Theoretical Prediction of Parallel Artificial Membrane Permeability Assay

Permeability using a Two-QSAR Approach

Max K. Leong

National Dong Hwa University, Taiwan

Abstract—Drug absorption is one of pivotal drug metabolism and

pharmacokinetics (DM/PK) properties that should be taken into account in

the process of drug discovery and development. The in vitro parallel

artificial membrane permeability assay (PAMPA) has been adopted as the

preliminary screening to estimate the passive diffusion of compounds in the

practical applications. A classical quantitative structure–activity relationship

(QSAR) model and a machine learning (ML)-based QSAR model were

derived using the partial least square (PLS) scheme and hierarchical support

vector regression (HSVR) scheme to elucidate the underlying passive

diffusion mechanism and to estimate the PAMPA effective permeability

(Pe), respectively, in this investigation. It was observed that HSVR

performed better than PLS as indicated by the predictions of the molecules

in the training set, test set, and outlier set as well as various statistical

assessments. When applied to the mock test, HSVR also demonstrated better

predictive performance. PLS, conversely, can render some relationships

between descriptors and permeability. Thus, the synergy of predictive

HSVR and interpretable PLS models can be greatly useful in facilitating

drug discovery and development by predicting passive diffusion.

15:00-16:00 Coffee Break & Poster Session

Page 35: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 35 -

Session 3

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 20, 2020 (Monday)

Time: 16:00-17:45

Venue: Room 1 (1F)

Topic: “Biological Science and Information Technology”

Session Chair: Prof. Kuo-Sheng Cheng

K1007

Session 3

Presentation 1

(16:00-16:15)

Effects of 2, 3', 4, 4', 5-pentachlorobiphenyl Exposure during Pregnancy dn

Germ Cell Development and Epigenetic Modification of F1 Mice

Qi-Long He, Xu-Yu Wei, Qian Zhou, Hai-Quan Wang and Shu-Zhen Liu

Shandong Normal University, China

Abstract—2, 3', 4, 4', 5-pentachlorobiphenyl (PCB118) is an important

dioxin-like polychlorinated biphenyls compound with strong toxicity.

PCB118 can accumulate in adipose tissue, serum and milk in mammals, and

it is highly enriched in the follicular fluid. In this study, pregnant mice were

exposed to 0, 20 and 100 μg/kg/day of PCB118 during pregnancy at the fetal

primordial germ cell migration stage. The methylation patterns of some

imprinted genes as well as the expression levels of Dnmts, Uhrf1 and Tets in

fully grown oocytes, sperm or testes and the methylation level in oocytes or

testes were measured in offspring. The rates of in vitro maturation, in vitro

fertilization, oocyte spindle and chromosomal abnormalities were also

calculated. The results showed that prenatal exposure to PCB118 altered the

DNA methylation status of differentially methylated regions in some

imprinted genes, the methylation levels in oocytes and testes also reduced.

The expression levels of Dnmts, Uhrf1 and Tet3 were changed by prenatal

exposure of PCB118 in oocytes or testes. In addition, PCB118 disturbed the

development process of progeny mouse germ cells in a dose-dependent

manner. Therefore, attention should be paid to the potential impacts of

PCB118-contaminated dietary intake during pregnancy on the offspring‘s

reproductive health.

K2027

Session 3

Presentation 2

(16:15-16:30)

Morphological Variation and Systematic Value of Indumentum in

Philippine Vavaea Bentham (Meliaceae) Species

Dhiocel A. Celadiña and Edwino S. Fernando

Western Philippines University, Philippines

Abstract—There is only one recognized species of Vavaea in the

Philippines, Vavaea amicorum. Vavaea amicorum is widespread and

Page 36: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 36 -

polymorphic in nature having a very broad circumscription. However,

recent collections in Surigao, Philippines from two populations discovered

that this species can hyperaccumulate nickel while others cannot. Thus, the

very broad circumscription of the widespread Vavaea amicorum in the

taxonomic revision of Pennington (1969) needs a closer re-examination.

Leaf morphological variations and indumentum characters of the eight

Philippine Vavaea species previously recognized were studied. Description

of each taxon is presented, and dichotomous key is provided for quick

reference. Principal component and cluster analyses of all species was

performed, using 25 morphological characters, to test their morphological

distinctiveness. The Principal Component Analysis generated five axes

explaining 85.73% of the total variance among the species. The most

important morphological characters differentiating Philippine Vavaea are the

indumentum type on branchlet, branchlet tips, petiole, leaf surfaces, midrib

and lateral veins. Three species may be resurrected, namely: Vavaea

pilosa, Vavaea pachyphylla and Vavaea surigaoense based from cluster

analysis. These three distinct morphological entities may be assigned back

to its previous taxonomic rank. Correct taxonomic position of Philippine

Vavaea is essential in crafting their conservation measures.

K0043

Session 3

Presentation 3

(16:30-16:45)

Deep Learning based System to Extract Agricultural Workers‘ Physical

Timeline Data for Acceleration and Angular Velocity

Shinji Kawakura and Ryosuke Shibasaki

The University of Tokyo, Japan

Abstract—Several physical characteristics of workers can be extracted from

physical timeline data to understand acceleration and angular velocity.

Although various approaches have been implemented globally for indoor

and outdoor agricultural (agri-) working sites, there is room for

improvement. In this study, we aim to adapt these approaches particularly

for real agri-directors, leaders and managers to improve the quality of tasks

and their security levels. Thus, we apply a deep learning-based method and

qualitatively demonstrate the classification of physical timeline datasets. To

create our dataset, our subjects were six experienced agri-manual workers

and six completely inexperienced men. The targeted task was cultivating the

semi-crunching position using a simple, Japanese-style hoe. We captured the

subjects‘ acceleration and angular velocity data from an integrated

multi-sensor module mounted on a wood lilt 15 cm from the gripping

position of the dominant hand. We used Python code and recent distributed

libraries for computation. For data classification, we successively executed a

Recurrent Neural Network (RNN), which we evaluated using wavelet

analyses such as the Fast Fourier Transform (FFT). These methods of

analyzing digital data could be of practical use for providing key

suggestions to improve daily tasks.

K0013 A Method for the Inverse QSAR/QSPR based on Artificial Neural Networks

and Mixed Integer Linear Programming

Page 37: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 37 -

Session 3

Presentation 4

(16:45-17:00)

Rachaya Chiewvanichakorn, Chenxi Wang, Zhe Zhang, Aleksandar

Shurbevski, Hiroshi Nagamochi and Tatsuya Akutsu

Kyoto University, Japan

Abstract—In this study,we propose a novel method for the inverse

QSAR/QSPR. Given a set of chemical compounds G and their values a(G)

of a chemical property, we define a feature vector f (G) of each chemical

compound G. By using a set of feature vectors as training data, the first

phase of our method constructs a prediction function ψ with an artificial

neural network (ANN) so that ψ(f (G)) takes a value nearly equal to a(G) for

many chemical compounds G in the set. Given a target value a∗ of the

chemical property, the second phase infers a chemical structure G∗ such that

a(G∗) = a∗ in the following way. We compute a vector f ∗ such that ψ(f ∗) =

a∗, where finding such a vector f ∗ is formulated as a mixed integer linear

programming problem (MILP). Finally we generate a chemical structure G∗

such that f (G∗) = f ∗. For acyclic chemical compounds and some chemical

properties such as heat of formation, boiling point, and retention time, we

conducted some computational experiments with our method.

K2014

Session 3

Presentation 5

(17:00-17:15)

An Investigation of Factors Influencing Performance of RFID-based Vehicle

Detection

Li-Fei Chen, Ruei-Shiuan Tsai, Yuan-Jui Chang and Chao-Ton Su

National Tsing Hua University, Taiwan

Abstract—Electronic toll collection (ETC) is an automatic charge system.

The driver can pass through the toll booth without needing to stop via this

ETC system. The ETC system can help to lower the cost of charging and

reducing the travel time. In this presentation, we provide a real case study to

investigate the performance of the highway ETC system. The system uses

sensors to transmit radio waves and detect radio frequency identification

(RFID) tags that are attached to the car. Five machine learning techniques,

random forest, XG-boost, gradient boost, support vector machine, and neural

network, are applied to analyze ETC data for vehicle detection in order to

identify key features that affect RFID tag detection and increase RFID tag

detection accuracy rate. After implementation, several valuable insights are

collected to enhance RFID tag detection rate; therefore, company‘s

operating costs can be greatly reduced.

K2020

Session 3

Presentation 6

(17:15-17:30)

Energy Monitoring and Management System Design Based on Embedded

Structure

Horng-Lin Shieh, Cheng-Chien Kuo and Sheng-Bo Gao

St. John‘s University, Taiwan

Abstract—Due to the huge amount of energy demanded by humans, it

causes serious damage to the environment, for example, the destruction of

the ozone layer by the exhaust gas generated by thermal power plants. In

order to reduce energy waste, this study designed a smart energy monitoring

Page 38: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 38 -

system that combines ARM embedded system, wireless transmission and

database technologies. In the designed system, ARM Cortex-M4 was

adopted as the system core. The customer's power consumption data can be

transmitted to the back-end database via wire or wireless, and the power

consumption information can be instantly displayed in a graphical interface.

In order to save and reduce wasted energy, this study uses a sorting

algorithm to automatically unload loads according to the urgency of energy

demand to avoid unnecessary energy waste.

K0046

Session 3

Presentation 7

(17:30-17:45)

GpemDB: A Scalable Database Architecture with the Multi-omics

Entity-relationship Model for Integrating Heterogeneous Big-data in Precise

Crop Breeding

Liang Gong, Shengzhe Fan and Wei Wu

Shanghai Jiao Tong University, China

Abstract—With the development of high-throughput genome sequencing

and phenotype screening techniques, there is a possibility of leveraging

multi-omics to speed up the breeding process. However, heterogenous big

data complicate the progress and the lack of database supported end-to-end

association analysis impedes the efficient use of these data. In response to

this problem, a scalable entity relationship model and a database architecture

are proposed in this paper to manage the cross-platform data sets and

explore the relationship among multi-omics. First, the targeted omics

regarding to precise breeding of crops are clarified from the engineering

viewpoint of cultivars. Meanwhile, a typical breeding data content and

structure is demonstrated with the case study of rice (Oryza sativa L.) where

the agronomic phenotype traits are highlighted. Second, the breeding

multi-omics data structure, patterns and hierarchy are described with

entity-relationship modeling technique. Third, a general-purpose scalable

database, called GpemDB (integrating Genomics, Phenomics, Enviromics,

and management), is developed. The developed GpemDB, involving Gpem

metadata-level layer and informative-level layer, provides a visualized

scheme to display the content of the database and facilitates users to

manage, analyze and share the breeding data. GpemDB with application to

rice breeding demonstrates that the proposed database architecture and

models are feasible to serve as the foundation for precise breeding of crops.

Page 39: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 39 -

Session 4

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 20, 2020 (Monday)

Time: 16:00-17:45

Venue: Room 2 (1F)

Topic: “Environmental Chemistry and Materials”

Session Chair: Prof. Manuel Garcia Roig

K3002

Session 4

Presentation 1

(16:00-16:15)

Methotrexate Degradation by UV-C and UV-C/TiO2 Processes with and

without H2O2 Addition on Pilot Reactors

L. A. González-Burciaga, J. C. García-Prieto, C. M. Núñez-Núñez, M.

García-Roig, J. B. Proal-Nájera

University of Salamanca, Spain

Abstract—Methotrexate (MTX) is an anti-cancer drug that can be excreted

up to 90% after administration due to its low biodegradability. Advanced

Oxidation Processes (AOPs) are a feasible alternative for the elimination of

MTX in the environment. In this research, AOPs were performed in

specialized patented reactors (UBE Photocatalytic systems and BrightWater

Titanium Advanced Oxidation Process) under experimental pilot conditions.

Photolysis and heterogeneous photocatalysis (UV and UV/TiO2)

experiments were performed with and without addition of H2O2 and at

different initial pHs. Best degradation percentage was achieved by

photolysis when initial Ph was 3.5 and added H2O2 was 3 Mm, reaching a

MTX degradation of 82% after 120 min of reaction. HPLC-MS analysis of

the resulting samples showed four possible byproducts of MTX degradation,

which presented a higher ecotoxicity than the starting compound.

K2015

Session 4

Presentation 2

(16:15-16:30)

Empirical Approach for Modeling of Partition Coefficient on Lead

Concentrations in Riverine Sediment

Saadia Bouragba, Katsuaki Komai, and Keisuke Nakayama

Kitami Institute of Technology, Japan

Abstract—Since a large part of heavy metals input in aquatic system

accumulates in sediment, their concentrations in sediment are regarded as an

important indicator of the heavy metal pollution of aquatic environment.

The partition coefficient (Kd) is an empirical parameter that can represent

the interaction of heavy metals at the sediment-water interface in aquatic

system, however, it is not always stable with environmental conditions.

Therefore, the introduction of Kd model with dominant physicochemical

parameters would facilitate and improve the simulation of heavy metals

Page 40: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 40 -

concentrations in riverine sediment. The present study aims to develop a Kd

model considering four physicochemical properties in stream water in order

to simulate heavy metal concentrations in sediment of severely polluted

urban rivers. Lead (Pb) concentrations in sediment of Harrach River,

Algeria, were simulated using one-dimensional distributed hydrological

model incorporating with presented Kd model. Multivariable equation of Kd

model with physicochemical parameters (pH, suspended solid concentration

(SS), chemical oxygen demand (COD) and biological oxygen demand

(BOD)) was obtained from multiple regression analysis with observation

data in various environmental condition. Hydrological simulations were

tested with Kd model comparing to giving constant Kd. The numerical

results agreed better with Kd model than with constant Kd, where the results

accuracy increased from R2 0.05 for constant Kd to R

2 0.67 for Kd model.

K2023

Session 4

Presentation 3

(16:30-16:45)

Potassium Fluoride-doped CaMgAl Layered Double Hydroxides for HCl

Adsorption in the Presence of CO2 at Medium-high Temperature

Wei Wu, Tongwei Wang, Decheng Wang and Baosheng Jin

Southeast University, China

Abstract—In the present work, the potassium fluoride-doped CaMgAl

layered double hydroxides (CaMgAl-LDHs) were synthesized by

co-precipitation method, which were used as adsorbents for HCl adsorption

in a quartz reactor under medium-high temperature (400-800 ℃). In

addition, the influence of CO2 to the HCl removal capacity of Ca-Mg-Al

layered double oxides (CaMgAl-LDOs) was also investigated. The results

showed that the KF/CaMgAl-LDOs loading with 25wt.% KF was the

optimal adsorbent for HCl removal. The microstructure of the adsorbents

after reaction revealed that the adsorbents were encapsulated by dense

chloride, and the adsorption process was mainly dominated by chemical

adsorption, strong acid-base properties, specific surface area and mesopore

number. Furthermore, the moderate CO2 concentration (0~6%) in the flue

gas of the municipal solid-waste incinerators could lessen the HCl removal

capacity of the CaMgAl-LDOs. However, the reduction of the HCl removal

capacity was less than 9% when the CO2 concentration was below 10%.

Moreover, the most suitable reaction temperature was optimized as 600 ℃

for the CaMgAl-LDOs. It can be drawn that employing some anti-sintering

support was efficient to improve the absorption capacity of the

CaMgAl-LDOs.

K2025

Session 4

Presentation 4

(16:45-17:00)

Quantum Chemical Research on Mechanism of Hydrogen Chloride

Removal by Calcium Carbonate

Jie Zhu, Wei Wu and Dejian Shen

Hohai University, China

Abstract—A detailed quantum chemical calculation was performed to

explore the mechanism of hydrogen chloride removal by calcium carbonate.

The possible pathways of the reaction between hydrogen chloride and

Page 41: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 41 -

calcium carbonate were studied by using gaussian software. The nature of

bonding evolution based on the main reaction pathway was studied by using

both electron localization function and atoms in molecules analysis.

Moreover, the reaction equilibrium constant and the reaction rate constant

were calculated between 298 K and 1100 K based on the rate-determining

step of the main reaction pathway. The results showed that: (1) the main

pathway of the reaction between hydrogen chloride and calcium carbonate

was CaCO3+HCl→CaClCO2OH, CaClCO2OH+HCl→CaCl2·H2O·CO2,

CaCl2·H2O·CO2→CaCl2+H2O+CO2; (2) the activation energy barrier of the

main pathway was 61.00 kJ·mol-1

; (3) the reaction equilibrium constant or

the reaction rate constant decreased or increased with the increase of the

temperature, respectively. The present study may provide useful information

for understanding the reaction mechanism of calcium-based adsorbent.

K2024

Session 4

Presentation 5

(17:00-17:15)

Bifunctional MOFs as Efficient Catalysts for Catalytic Conversion of

Carbon Dioxide into Cyclic Carbonates and DFT Studies

Yuanfeng Wu and Guomin Xiao

Southeast University, China

Abstract—The metal organic framework materials of

[(CH3NH3][Mn(COOH)3] (MA-MnF), [(CH3CH2NH3][Mn(COOH)3]

(EA-MnF), and [C3H5N2][Mn(COOH)3] (Im-MnF) were easily prepared

under room temperature, and further employed as the highly-efficient

catalysts for catalytic conversion of carbon dioxide into cyclic carbonates.

The metal organic frameworks including MA-MnF, EA-MnF, Im-MnF were

studied via several characterizations such as XRD, FT-IR, XPS,

N2-adsorption, TG-DSC, CO2-adsorption and NH3-TPD. Interestingly,

Im-MnF compound was observed to possess the highest catalytic activity

among the studied compounds, which is associated not only with the

amounts of basic sites, but also related to the nitrogen-containing species.

Various reaction parameters including reaction temperature, reaction time,

catalyst loading, reaction pressure as well as recyclability and coupling

scope were explored. 97.27% conversion of allyl-glycidyl ether (AGE, TOF:

36.78 h-1

) and 97.66% selectivity to allyl-glycidyl carbonate (AGC) were

obtained under the explored optimized conditions (100 oC, 15bar, 6h, 1.0

wt.% of AGE). In addition, only a slight downward in catalytic activity was

found when the sample was reused twice. Furthermore, coupling reactions

of CO2 with various epoxides were also performed, of which, the yield of

the cyclic carbonates followed the order: Epichlorohydrin > Allyl glycidyl

ether > Styrene oxide > Cyclohexene oxide> Propylene oxide. Finally, a

mechanism was proposed for explaining carbon dioxide insertion into

epoxide. Besides density functional theory (DFT) calculations were also

performed on Gaussian 09 Packages (M06 method//6-31g(d, p) for

main-group element, ECP-Lanl2DZ for Mn element). The calculated results

were in good agreement with the DFT calculation.

K1013 A Novel Deep-blue Thermally Activated Delayed Fluorescence Emitter

Page 42: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 42 -

Session 4

Presentation 6

(17:15-17:30)

Dimethylacridine/Thioxanthene-S,S-Dioxide for Highly-efficient Organic

Light-emitting Devices

Young Pyo Jeon

Hanyang University, South Korea

Abstract—Even though the development of organic light-emitting devices

(OLEDs) as a potential sustainable future display technology has made great

progress, a study on materials with deep-blue emission for use in thermally

activated delayed fluorescence (TADF) OLEDs is still an important

challenge in this field of scientific research. Hence, a novel deep-blue

emissive dopant material,

2,7-bis(9,9-dimethylacridine-10(9H)-yl)-9,9-diphenyl-9H-thioxanthene

10-,10-dioxide (DMA-ThX), is synthesized for highly-efficient thermally

activated delayed fluorescence organic light-emitting devices (TADF

OLEDs). The molecular design of DMA-ThX retains the overlap between

the lowest singlet excited state (S1) and the lowest triplet excited state (T1)

because of its twisted molecular structure. From photoluminescence

emission measurements at low temperature, the difference between of the S1

and T1 energy levels of 3.14 and 3.07 eV is found to be 0.07 eV.

Furthermore, the external quantum efficiency of the deep-blue-emitting

TADF OLEDs with DMA-ThX-doped bis[2-(diphenylphosphino)phenyl]

ether oxide is 18.4%, which is significant, and the

Commission Internationale de l'Eclairage coordinates are (0.14, 0.14),

indicative of very pure deep-blue emission.

K2036

Session 4

Presentation 7

(17:30-17:45)

Environmental Sustainability: Management Perception in Oil and Gas

Industry in Libya

Nahg A. Alawi, Khairi Ahmed Masaud and Samer A. Bamansoor

Aden University, Yemen & Geomatika University, Malaysia

Abstract—This study aims to examine the environmental sustainability

perceptions of corporate managers in oil and gas industry in Libya and the

factors that effect this perception. Based on primary data collected via a

self-administered survey sent to 274 managerial and executive level and

analysed using descriptive means and regression. The result shows that

managers have high level of environmental sustainability perception. The

results also show that only educational level have a positive and significant

impact on managers‘ environmental sustainability perceptions. Limitation of

this study is noted including the generalizability of the findings within

organisation in Libya. The study has bridged the literature gaps in such that

it offers empirical evidence and new insights on environmental

sustainability body of knowledge which could be used to further

improvement of the environmental sustainability perception in oil and gas

companies in Libya.

Page 43: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 43 -

Session 5

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 21, 2020 (Tuesday)

Time: 13:30-15:15

Venue: Room 1 (1F)

Topic: “Bioinformatics”

Session Chair: Prof. Tomohiro Araki

K0077

Session 5

Presentation 1

(13:30-13:45)

A Sparse Bayesian Approach to Combinatorial Feature Selection and Its

Applications to Biological Data

Ryoichiro Yafune, Daisuke Sakuma, Yasuo Tabei, Noritaka Saito, Einoshin

Suzuki and Hiroto Saigo

Kyushu University, Japan

Abstract—With the rapid increase in the availability of large amount of

data, prediction is becoming increasingly popular, and has widespread

through our daily life. However, powerful non-linear prediction methods

such as neural networks and SVM suffer from interpretability problem,

making it hard to use in domains where the reason for decision making is

required. In this presentation, we present a prediction model that consider

variable interactions, in which salient variables are selected based on a

Bayesian sparse prior. We demonstrate its usefulness in bioinformatics

applications, such as detecting interactions among biomarkers.

K0017

Session 5

Presentation 2

(13:45-14:00)

Analyses of Interaction between Platinum Bonded LARFH and Gold

Surface by Molecular Dynamics Simulation

Mao Watabe, Keiichi Nobuoka, Hironao Yamada, Takeshi Miyakawa, Ryota

Morikawa, Masako Takasu, Tatsuya Uchida and Akihiko Yamagishi

Tokyo University of Pharmacy and Life Sciences, Japan

Abstract—Proteins that specifically bind to metals have been used for

research on development of new organic-inorganic hybrid materials. Several

peptides and proteins that bind to metals have been reported; this property

can be attributed to their structures. In this paper, we assessed the binding of

a sequence of the protein called LARFH to a metal surface. We performed

simulation with several models of LARFH protein with different loop

sequences binding to the metal surface. It was found that the angle between

the LARFH protein and the metal surface varied for different loop

sequences. We also performed simulations of the detachment process of

LARFH. It was concluded that the stronger binding to metal is established

by two gold-binding loops of the LARFH sequences.

Page 44: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 44 -

K0038

Session 5

Presentation 3

(14:00-14:15)

Identification of Redox-sensitive Cysteines using Sequence Profile

Information

Md. Mehedi Hasan and Hiroyuki Kurata

Kyushu Institute of Technology, Japan

Abstract—Redox-sensitive cysteine (RSC) thiol contributes to many

biological processes. The identification of RSC plays an important role in

clarifying the mechanism of redox-sensitive factors; nonetheless,

experimental investigation of RSCs through traditional methods is

expensive and time-consuming. Therefore, the computational approaches

that potentially predict candidate RSCs using the sequence information are

critically needed. Herein, a sophisticated computational predictor IRC

(Identification of Redox-sensitive Cysteine) was developed to identify the

RSC using sequence profile information with a non-parametric feature

selection approach. Then, a machine learning algorithm called ―random

forest‖ was trained with these features to build the predictor. The resultant

IRC achieved an AUC score of 0.803 on the training dataset. The IRC

performed better than other existing computational models on both the

training and independent datasets. The IRC is a helpful computational

predictor for the prediction of RSCs..

K0019

Session 5

Presentation 4

(14:15-14:30)

Relationship between Dynamics of Structures and Dynamics of Hydrogen

Bonds in Hras-GTP/GDP Complex

Takeshi Miyakawa, Kimikazu Sugimori, Kazutomo Kawaguchi, Masako

Takasu, Hidemi Nagao and Ryota Morikawa

Tokyo University of Pharmacy and Life Sciences, Japan

Abstract—Hras protein is an intermediate for signals of cell proliferation

and cell differentiation when Hras combines with guanosine triphosphate

(GTP). In ordinary cells, GTP combined with Hras is hydrolyzed to

guanosine diphosphate (GDP), and the structures of the protein-ligand

complex strongly depend on hydrogen bonds. In such a system, hydrogen

bonds exist between protein and ligand, between protein and water, and

between ligand and water. In this study, we applied a relaxation mode

analysis (RMA) to the trajectories of MD simulations for Hras-GTP/GDP to

investigate the relationship between the dynamics of structures and those of

hydrogen bonds. Different relaxation time of relaxation modes of Hras

structures and Hras-solvent water hydrogen bonds were evident between

Hras-GTP and Hras-GDP, and the results imply that the structure of Hras in

Hras-GTP is stiffer than that in Hras-GDP. The method devised here and its

interpretation can be applied to other systems focusing on objects with

different stiffnesses.

K0079

Session 5

Bayesian Optimization for Sequence Data

Kohei Oyamada and Hiroto Saigo

Kyushu University, Japan

Page 45: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 45 -

Presentation 5

(14:30-14:45)

Abstract—Directed evolution is a method used in protein engineering that

mimics the process of natural selection to steer proteins or nucleic acids

toward a user-defined goal. It iteratively introduce randomness in sequence,

and choose members with desired functionalities. In order to decrease the

number of experimental rounds, we consider introducing a strategy known

as a Bayesian optimization. It requires similarity measure among sequences,

so we employ vectorization methods such as k-mers and seq2seq as well as

alignment scores. We validate the effectiveness of our approach in four

benchmark datasets. In the two tasks where consideration of evolutionary

distance is deemed to be important, alignment scores showed better

performance than vectorization methods. We also demonstrate the

usefulness of introducing batch processing for accelerating the computation.

K0020

Session 5

Presentation 6

(14:45-15:00)

Microorganism Hydrolysate Reduced Lipo-oxidation in Regulated

High-fructose Induced Nonalcoholic Fatty

Liver Disease in A Murine Model

Chang-Chi Hsieh and Li-Jia Huang

Tunghai University, Taichung, Taiwan

Abstract—Pet suffered from metabolic syndrome were change in eating habits

in diet. In Taiwan, Sixty percent of pets have obesity problems, and the

obese pets are at higher risk of nonalcoholic fatty liver disease (NAFLD).

Pig kidney is a large amount of slaughter by-products. We developed the

pig kidney by-products were recycled and applied to healthy food for pets.

The purpose of a study, we will use Lactobacillus Plantarum hydrolyzed pig

kidney by-products, and explored the effects of hydrolysate on NAFLD in

murine model. In this study, male C57BL/6JNarl mice were divided into

five groups, normal group, control group (administered with the distilled

water), treatment group (administered with pig kidney by-products

microorganism hydrolysate 50 and 200 mg/kg/day) and microorganism

control group. And mice induced NAFLD by 30% high-fructose corn

syrup solution for 8 weeks. Finally, we measured serum aspartate

aminotransferase (sAST), alanine aminotransferase (sALT), triglyceride

(sTG), total cholesterol (sTC) and cytokines (IL-6, MCP-1, adiponectin).

Pathology of liver staining sections were determinate by H&E staining and

immunohistochemical staining for 4 Hydroxynonenal (4-HNE), thymic

stromal lymphopoietin (TSLP). The results indicate the microorganism

hydrolysate decreased sTG, sTC, IL-6, MCP-1 and increased adiponectin.

The inhibition of lipo-oxidation might be the results of the improve in

NASLD by reduced the lipo-oxidation related indicators.

K0045

Session 5

Presentation 7

(15:00-15:15)

Improvement of Protein Stability Prediction by Integrated Computational

Approach

Chi-Wei Chen, Meng-Han Lin, Hsung-Pin Chang and Yen-Wei Chu

National Chung-Hsing University, Taiwan

Abstract—Mutation of a single amino acid residue may change protein

Page 46: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 46 -

structure which affect protein function and diseases. Increasing protein

stability or maintaining its stability while changing protein properties is

often a goal in protein engineering, drug design or industrial optimization. A

variety of methods and features have been proposed to predict the stability

of protein mutations, the conflicting prediction results from different tools

could cause confusion to users. Therefore, this study integrates structure and

sequence-based prediction tools with machine learning and adds information

of protein sequences. The best model is selected through feature selection

methods to improve accuracy and reduce the time complexity of the training

model. The PCC (Pearson correlation coefficient) of the integrated model

can be increased from 0.669 to 0.702. Not only successfully integrates

predictors, but also improves the accuracy of integration tools.

Page 47: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 47 -

Session 6

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 21, 2020 (Tuesday)

Time: 13:30-15:15

Venue: Room 2 (1F)

Topic: “Biomedicine”

Session Chair: Prof. Hongmin Cai

K0034

Session 6

Presentation 1

(13:30-13:45)

Expression of TLR4 Signaling in Acute Appendicitis and Experimental

Peritonitis

Han-Ping Wu

China Medical University Children's Hospital, Taiwan

Abstract—Background: Simple appendicitis can progress to perforation,

which may carry septic peritonitissepsis, and mortality. Toll-like receptors

(TLRs) and myeloid differentiation factor 88 (MyD88) are essential for

pathogen recognition and protective immunity. Methods: From 2015 to

2016, 126 pediatric patients with suspected appendicitis were enrolled. The

protein expressions in excised appendices were analyzed by

immunohistochemistry (IHC) staining. An animal study was investigated in

rats with septic peritonitis induced by cecal ligation and puncture (CLP).

The expressions of MyD88-dependent pathway biomarkers, including

MyD88, nuclear factor-κB (NF-κB) and serum tumor necrosis factor-α

(TNF-α) were analyzed and compared to the sham controls at the different

time points after CLP surgery. Results: MyD88 expression highly increased

in early-stage simple appendices. In our animal model, CLP-induced sepsis

increased liver MyD88 mRNA and protein expressions at 2 hours after

surgery. MyD88 protein expressions in rats with CLP-induced sepsis marked

increased at 4 and 6 hours, and their NF-κB activities and serum TNF-α

levels also increased at 4 hours after CLP surgery (both p<0.05).

Conclusion: This study showed different expressions of proinflammatory

markers in pediatric appendicitis. The different serial expression of

MyD88-dependent pathway during sepsis may serve as biomarkers during

sepsis.

K0049

Session 6

Presentation 2

(13:45-14:00)

Mapping the Humoral Immune Response of Patients with American

Trypanosomiasis using Phage Display Technology

André A. R. Teixeira, Luis A. Rodriguez-Carnero, Edécio Cunha-Neto,

Walter Colli and Ricardo J. Giordano

University of São Paulo, Brazil

Page 48: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 48 -

Abstract—Trypanosomiasis is a neglected disease caused by Trypanosoma

cruzi. It is known that the immune system plays important roles in the

disease, but its molecular mechanisms are still unresolved. Besides, it is

unknown why just 30% of patients eventually develop cardiomyopathy,

while most remain asymptomatic. We need better molecular markers for

diagnosis, for following disease progression, and for cure. Considering all

this, we produced a comprehensive T. cruzi Genomic Shotgun phage display

library and used it to select epitopes recognized by IgG from patients in

different clinical stages of the disease. The library we built had coverage of

at least 120 times the T. cruzi genome and allowed us to identify more than

400 epitopes. Many of them had never been associated with the disease and

are still considered hypothetical proteins with unknown functions. ELISA

assay was used to validate 8 epitopes identified in our study, confirming our

findings. It also revealed that individual patient response is very

heterogeneous, adding another layer of complexity to the disease. In sum,

our findings might be important to identify new markers for disease

progression and cure. Also, it may work as blueprint for antigenic profiling

of patients with other neglected diseases.

K1009

Session 6

Presentation 3

(14:00-14:15)

External Harmful Substances Induced Sensitization and Its Association

Assessments of Inflammation Mechanisms in Epithelial Cell and Animal

Model

Zhi-Yao Ke, Thung-Shen Lai and En-Chih Liao

Mackay Medical College, Taiwan

Abstract—Environmental pollutants are compounds introduced in the

natural environment causing adverse changes, for example, adversely

affecting health or causing other types of damage. The house dust mites are

the most important sources of indoor allergens responsible for the

development of asthma. Staphylococcal enterotoxin B (SEB), is

an enterotoxin produced by the Staphylococcus aureus, common cause

of food poisoning. Incense burning is an integral part of daily lives in large

parts of Asia. To build a cell and animal model of multi-environmental

factors(House dust mites, staphylococcal enterotoxin B and incense smoke)

induced inflammation, we can proof its feasibility by study the correlation

between monocyte associated inflammasome activation and the production

of inflammatory cytokines. IgE and IgG1 in sera were increased

significantly when compared with those of the normal saline group after

sensitization. Splenocyte were collected and cultured with PMA and

ionomycin for 5 hours before CD4+ cells producing the TH1-type cytokine

IFN-a and TH2-type cytokine IL-4 were analyzed the percentage of CD4

cells producing IL-4 in HDM-sensitized mice was higher than the normal

saline group. Lung tissue was acquired for polymerase chain reaction assay

to investigate gene expression, including T-bet, GATA-3, IL33, the

proinflammatory gene IL6. The results of polymerase chain reaction

revealed significant upregulation of the gene expression after sensitization in

Page 49: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 49 -

lung. The cytokine productions (IL4, IL6, and TNF-α) were significantly

increased compared with the normal saline group after sensitization. The

findings indicated that sensitization with external harmful substances led to

significant expression of lung tissue inflammation, tracheal thickening, and

tracheal rupture. In conclusion, the effects of different environmental

factors(House dust mites, staphylococcal enterotoxin B and incense smoke)

may have a synergistic effect in the inflammation reactions.

K2017

Session 6

Presentation 4

(14:15-14:30)

The Molecular Mechanism and Drug Therapy of Heart Failure

Yifan Su

SUNY-ESF, USA

Abstract—Heart failure (HF) is a complex clinical syndrome that results

from left ventricular myocardial dysfunction and contributes to dyspnea,

fatigue and fluid retention. It's essential to characterize disease progression

and symptoms to optimize therapy selection. Also, understanding the

mechanisms of HF to guide therapy selection is of vital importance. This

review focuses on the demonstrating mechanisms of HF and points out the

possible pathways involved in the process. Drug therapies of HF with

different effects on specific targets are analyzed. Finally, we conclude the

features of the now going drugs and depict the future perspectives of drug

therapy of HF, including potential new targets and new drug therapies.

K2003

Session 6

Presentation 5

(14:30-14:45)

Rapid Dysphagia Diagnostic Test based on Voice Features

Hsin-Yu Chou, Li-Ya Lin, Li-Chun Hsieh, Pei-Yi Wang and Shih-Tsang Tang

Ming Chuan University, Taiwan

Abstract—The diagnosis of dysphagia is quite time consuming and

uncomfortable. Most patients usually do not seek medical treatment until the

situation deteriorates, which results in missing the golden treatment period.

Because the organs involved in swallowing and vocalization are almost

overlapping, both of their function may decline simultaneously. Therefore,

the aim of this study is to develop the rapid dysphagia diagnostic test based

on voice features. In the clinical trial, the subject was firstly evaluated by the

swallowing questionnaire scales, and then their voices were recorded. After

extraction of the voice features, the correlation between swallowing scales

and voice features were analyzed finally. A total of 40 subjects had

completed the trial. The results showed that the dysphagia severity was

highly correlated with jitter, shimmer, HNR and speech rate significantly.

For the patient, the voice features analysis is easily carried out, which only

requires vocalization. The record of voice is almost unrestricted by time and

place. Hence rapid dysphagia diagnostic test would be clinically feasible,

and it prompts early diagnosis and early treatment.

K0059

Session 6

Corpus Construction of Precision Medicine

Xuejing Ren, Xinying An and Shaoping Fan

Chinese Academy of Medical Sciences, China

Page 50: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 50 -

Presentation 6

(14:45-15:00)

Abstract—[Background] For advancing biomedical text-mining research,

formal evaluations and manually annotated text corpus are critically

important. In terms of biomedical corpus construction, in order to meet

different needs, many scholars have built different biomedical corpus with

different emphasis. Through horizontal observation and vertical comparison

of existing biomedical corpus, it is found that most of the entities of

biomedical corpus are roughly classified and have narrow coverage. Most of

the classification and sub-classification of entities only includes genes,

proteins, drugs, diseases and other entities, while the classification of

pathways, mutations and other entities is rarely involved. The exploration of

the relationship between entities is not enough in depth and the relationship

type is uncomprehensive. Therefore, we have developed our own precise

medical corpus to store more comprehensive medical knowledge and enrich

the biomedical corpus.[Methods]Through repeated iteration the process of

article selection, systematic evaluation, manual annotation, formulating

consensus of annotation, cross validation, submission assessment to form a

more comprehensive and authoritative text corpus, which is also of great

significance to promote biomedical text mining research.[Results]At

present, a total of 6000 articles have been auto and manual annotated. Each

1000 PubMed English articles of cardiovascular diseases (circulatory

system) intestinal neoplasm and liver neoplasm (digestive system) metabolic

diseases; 1500 articles of lung neoplasm (respiratory system) and 500

articles of neurological diseases. (but in this issue, we only had a detailed

illustration of the lung neoplasm.[Conclusion]A high-quality precise

medical corpus was constructed, which promotes the research process of

biomedical data mining. At present, some resources of the corpus have been

opened to the public, welcome scholars to use and put forward valuable

suggestions.

K0040

Session 6

Presentation 7

(15:00-15:15)

Computational Drug-target Interaction Prediction based on Graph

Embedding and Graph Mining

Maha A. Thafar, Somayah Albaradie, Rawan S. Olayan, Haitham Ashoor,

Magbubah Essack and Vladimir B. Bajic

King Abdullah University of Science and Technology (KAUST), Saudi

Arabia

Abstract—Identification of interactions of drugs and proteins is an essential

step in the early stages of drug discovery and in finding new drug uses.

Traditional experimental identification and validation of these interactions

are still time-consuming, expensive, and do not have a high success rate. To

improve this identification process, development of computational methods

to predict and rank likely drug-target interactions (DTI) with minimum error

rate would be of great help. In this work, we propose a computational

method for (Drug-Target interaction prediction using Graph Embedding and

graph Mining), DTiGEM. DTiGEM models identify novel DTIs as a link

prediction problem in a heterogeneous graph constructed by integrating

Page 51: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 51 -

three networks, namely: drug-drug similarity, target-target similarity, and

known DTIs. DTiGEM combines different techniques, including graph

embeddings (e.g., node2vec), graph mining (e.g., path scores between drugs

and targets), and machine learning (e.g., different classifiers). DTiGEM

achieves improvement in the prediction performance compared to other

state-of-the-art methods for computational prediction of DTIs on four

benchmark datasets in terms of area under precision-recall curve (AUPR).

Specifically, we demonstrate that based on the average AUPR score across

all benchmark datasets, DTiGEM achieves the highest average AUPR value

(0.831), thus reducing the prediction error by 22.4% relative to the

second-best performing method in the comparison.

15:15-16:15 Coffee Break & Poster Session

Page 52: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 52 -

Session 7

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 21, 2020 (Tuesday)

Time: 16:15-18:15

Venue: Room 1 (1F)

Topic: “Medical Informatics”

Session Chair: Prof. Chang-Chi Hsieh & Prof. Jiann-Shing Shieh

K0012

Session 7

Presentation 1

(16:15-16:30)

Rapid Detection and Prediction of Influenza A Subtype using Deep

Convolutional Neural Network based Ensemble Learning

Yu Wang, Junpeng Bao, Jianqiang Du and YongFeng Li

Xi‘an Jiaotong University, China

Abstract—Seasonal pandemics of influenza A viruses bring enormous

threaten to human healthy. Different subtypes of influenza A viruses

disseminated in human have variable susceptibilities to antiviral drug, so

rapid subtyping of influenza A viruses has been increasingly important.

Traditional biochemical methods for subtyping these viruses are expensive

and time-consuming. Various sequencing techniques and deep learning

methods bring an opportunity to analyse and gain information of those biont

more conveniently and accurately. This paper proposes a deep convolutional

neural network based ensemble learning model to precisely detect all

subtypes of influenza A viruses. The experiments show that the proposed

method can achieve the state-of-art performance for subtyping influenza A

viruses and detecting a fire-new subtypes according to sequence data.

K0015

Session 7

Presentation 2

(16:30-16:45)

Predicting lncRNA-disease Association based on Extreme Gradient

Boosting

Xi Tang, Menglu Li, Wei Zhang and Junfeng Xia

Anhui University, China

Abstract—There is increasing evidence that long non-coding RNAs

(lncRNAs) play an important role in many significant biological processes.

Associations‘ detection between lncRNAs and human diseases by

computational models is beneficial to the identification of biomarkers and

the discovery of drugs for the diagnosis, treatment, and prognosis of human

diseases. In this study, we propose a method called PrLDA (Predicting

LncRNA-Disease Association based on extreme gradient boosting) for

predicting potential lncRNA-disease associations based on eXtreme

Gradient Boosting (XGBoost). Firstly, we compute semantic similarity of

diseases and lncRNA sequence similarity. Then, we extracte feature vectors

Page 53: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 53 -

by concatenating these similarities horizontally. At last, the feature matrix

after dimension reduction is used as the input for XGBoost and we get the

score about the lncRNA association with a specific disease. Computational

results indicate that our method can predict lncRNA-disease associations

with higher accuracy compared with previous methods. Furthermore, case

study shows that our method can effectively predict candidate lncRNAs for

breast cancer, with 80% of the top 10 predictions are confirmed by

experiments. Therefore, PrLDA is a useful computational method for

lncRNA-disease association prediction.

K0041

Session 7

Presentation 3

(16:45-17:00)

Breast Cancer Subtype by Imbalanced Omics Data through A Deep

Learning Fusion Model

Jingwen Zeng, Hongmin Cai and Tatsuya Akutsu

South China University of Technology, China

Abstract—Breast cancer is a highly heterogeneous disease that consists of

subtypes with distinct genetic features and clinical symptoms. The patients

with different subtypes react to different therapies, thus identifying

molecular subtypes greatly contributes to precision diagnosis and

personalized cancer treatment. PAM50 subtype is a widely accepted

standard in breast cancer classification. The large amount of multi-omics

data in public database like TCGA greatly contribute to the study of breast

cancer subtype identification. However, the imbalance of sample subtypes in

the existing database results in a large difficulty in correctly identifying

subtypes with small sample size. In this paper, we proposed a novel method

to accurately identify the PAM50 subtypes by utilizing the patients‘ omics

profiles in TCGA database. Based on the integrated expression profiles of

RNA-seq and Copy Number Alteration (CNA), the proposed method

identifies subtype-related patterns by a multi-layer Convolutional Neural

Network (CNN). A weighted loss function was applied to alleviate the

effects of imbalanced samples, thus contributing to the accurate

identification. We demonstrated that our method could identify PAM50

subtypes of patients with high precision (90.02%) and outperformed two

benchmark methods.

K2006

Session 7

Presentation 4

(17:00-17:15)

Design and Implementation of NoSQL-based Medical Image Archive

Underlaying FHIR

Chien-Hua Chu, Shih-Tsang Tang and Chung-Yueh Lien

Ming Chuan University, Taiwan

Abstract—In this study, we proposed the architecture of the NoSQL-based

medical image archive underlying Fast Healthcare Interoperability

Resources (FHIR) to provide several web services for medical image

retrieval applications. The Node.js framework is used to create an HTTP

server providing web services for medical image retrieval including CRUD

operations (create, read, update and delete) to manage the DICOM (Digital

Imaging and Communications in Medicine) objects via HTTP protocol. A

Page 54: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 54 -

set of DICOM objects, uploaded to the server, are processed and converted

to JSON formatted documents stored in a NoSQL database with MongoDB.

To maintain the consistency of the hierarchical relationship among the

DICOM objects, the schema of patient-based architecture combined with

FHIR ImagingStudy was determined according to DICOM hierarchical data

structure, i.e. patient/study/series/objects. The development of the

NoSQL-based method does not only improve the shortcomings of traditional

SQL-based image archive but also supports the FHIR standard that

effectively improves the efficiency of web-based medical retrieval.

K0066

Session 7

Presentation 5

(17:15-17:30)

A Novel Feature Selection and Classification Method of Alzheimer's

Disease based on Multi-features in MRI

Peiqi Luo, Guixia Kang and Xin Xu

Beijing University of Posts and Telecommunications, China

Abstract—In this paper, we describe a novel machine learning method for

classifying Alzheimer‘s disease (AD), Mild cognitive impairment (MCI) and

Normal Control (NC) subjects based on structural MRI. We first extracted

features from MRI scans, including cortical volumes, cortical thicknesses,

subcortical volumes, and hippocampal subfields volumes. Then a new

feature selection method combining the support vector machine-recursive

feature elimination (SVM-RFE), maximal-relevance-minimal-redundancy

(mRMR) and random forest (RF) was proposed to select the optimal subsets

among all these features. Finally, the SVM classifier was used for

AD/MCI/NC classification by 10-fold cross-validation. We applied the

proposed method to the Alzheimer‘s Disease Neuroimaging Initiative

(ADNI) dataset, and the experimental results show a high degree of

accuracy, sensitivity and specificity, which are superior to some other

state-of-the-art approaches

K0071

Session 7

Presentation 6

(17:30-17:45)

Tracking and Analysis of FUCCI-labeled Cells based on Particle Filters and

Time-to-event Analysis

Kenji Fujimoto, Shigeto Seno, Hironori Shigeta, Tomohiro Mashita,

Junichi Kikuta, Masaru Ishii and Hideo Matsuda

Osaka University, Japan

Abstract—FUCCI (fluorescent ubiquitination-based cell cycle indicator) is a

fluorescent probe used to visualize the cell cycle progression of individual

cells using fluorescent proteins of different colors. Because the cell cycle is

related to biological processes such as proliferation of cancer cells, analysis

of imaging data visualized using FUCCI is extremely important. This paper

proposes a method for spatiotemporal tracking and analysis of

FUCCI-labeled cells from time-lapse videos. To address the color transition

of the FUCCI-labeled cell with the cell cycle progression, the proposed

method simultaneously estimates the location and the cell cycle phase of the

target cell. Furthermore, to analyze the cell phase transition, this paper

proposes to apply multistate time-to-event analysis to the information

Page 55: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 55 -

obtained through our tracking method. This paper demonstrates the

usefulness of our method with application to FUCCI-labeled HuH7 cells

(human hepatocellular carcinoma cell line).

K2004

Session 7

Presentation 7

(17:45-18:00)

Home Wound Dressing Photo Management Cloud System

Meng-Jie Su, Yu-Tong Liu, Szu-Hsien Wu, Ming-Liang Hisiao and

Shih-Tsang Tang

Ming Chuan University, Taiwan

Abstract—Many patients with chronic trauma often need wound dressings

by themselves. The wound dressing affect directly the wound‘s healing.

Therefore the quality of home wound dressing becomes very important. This

study developed a home wound dressing photo management cloud system,

which is to collect home wound dressing photos, and then evaluate the

wound condition. The system would assist physicians in clinical evaluation

and patient‘s care recommendations. The proposed system developed a

series of interactive web forms basing on Windows operating system,

Apache web server and MySQL database engine. The wound dressing

photos are stored and managed by a cloud system. Patient could upload

wound dressing photos on a daily basis, which includes the photos before

the wound dressing and after the dressing. The system is further developing

AI functions, which would provide the trend forecasting in the wound

healing.

K0076

Session 7

Presentation 8

(18:00-18:15)

Development of Photo-activatable Insulin Receptor for Insulin Signalling

Pathway Control

Bryan John J. Subong, Mizuki Endo and Takeaki Ozawa

The University of Tokyo, Japan

Abstract—The insulin signaling pathway is one of the most crucial

biochemical pathways which controls energy functions such as glucose and

lipid metabolism. Impairment of the said signaling pathway has been

implicated in various diseases such as diabetes mellitus and cancer. Current

approaches in studying the said pathway uses mass-spectrometry driven

phosphoproteomics. However, this technique is destructive and does not

allow a spatio-temporal analysis of the said signaling cascade. In our present

study, we develop an optogenetic probe which can be used to study the

insulin signaling pathway spatio-temporally. The probe utilizes light for

dimerization of the insulin receptor beta-subunit instead of insulin. Upon

10-minute light irradiation, activation of various insulin signaling

downstream proteins such as phopho-irs1, phospho-Akt and phospho-Erk

were confirmed using western blotting. Further, glucose uptake assay using

luminescence technology to measure uptake of 2-deoxyglucose-6-phosphate

showed statistically significant (at alpha = 0.05) increased glucose uptake

response for cells transiently expressing the said probe. Further, the

steady-state localization of the probe expressing enhanced green fluorescent

protein was determined to be at the plasma membrane. Moreover, cell

Page 56: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 56 -

viability assay showed no cytotoxic effect between light irradiated and

non-light irradiated cells. Thus, the developed optogenetic tool can be safely

used to interrogate living systems. This study highlights the use of emerging

tools such as optogenetics in controlling various intracellular processes at a

spatio-temporal level. In particular, the development of the said probe will

allow better understanding of the effects of regulating and controlling

spatio-temporally the insulin signaling pathway in living systems.

Page 57: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 57 -

Session 8

Tips: The schedule for each presentation is for reference only. In order not to miss your presentation,

we strongly suggest that you attend the whole session.

Afternoon, January 21, 2020 (Tuesday)

Time: 16:15-18:15

Venue: Room 2 (1F)

Topic: “Preventive Medicine and Rehabilitation Medicine”

Session Chair: Prof. Han-Ping Wu

K3003

Session 8

Presentation 1

(16:15-16:30)

Variability of Local Weather as Early Warning for Dengue Hemorrhagic

Fever Outbreak in Indonesia

Khaidar Ali, Isa Ma‘rufi, Wiranto and Anis Fuad

Universitas Gadjah Mada, Indonesia

Abstract—The incidence of Dengue Hemorrhagic Fever (DHF) is related to

the alternation of environment condition, particularly weather, in which

global warming may elevate the DHF case. The objective of study is to

analyses the relationship between local weather and DHF, and to create

prediction model by using big data in Surabaya Municipally. Employing

quantitative method, monthly time series data was used during 2012-2016.

Univariate, bivariate and multivariate analysis was performed in Stata 13.

Local weather (mean humidity, maximum humidity, minimum humidity,

rainy days and rainfall) correlates to the incidence of DHF (p <0.05), in

which minimum humidity lag 1 month and rainy days lag 2 month had

strong correlation (r >0.7). In addition, prediction model in the study

recognize the occurrence of four peak epidemic of DHF cases in Surabaya

Municipally. Therefore, utilizing local weather to create prediction model

may contribute as early warning for DHF incident in order to handle the

disease in Surabaya Municipally.

K2035

Session 8

Presentation 2

(16:30-16:45)

Associations Between Seasonal Variation of Heart Rate Variability and

Healthy Life Expectancy in Japan

Emi Yuda, Masaya Kisohara, Yutaka Yoshida, Norihiro Ueda and Junichiro

Hayano

Tohoku University & Nagoya City University, Japan

Abstract—Although both summer heat and winter cold increase the risk of

health problems, the effects could be modified by living environment and

clothes. We investigated the relationship between regional differences in

healthy life span (HALE) and seasonal variations in heart rate variability

(HRV) that may reflect biological burden caused by heat and cold weather.

Using the Allostatic State Mapping by Ambulatory ECG Repository

Page 58: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 58 -

(ALLSTAR) database, we analyzed the 24-h standard deviation of

sinus-rhythm R-R intervals (SDNN) among 86,712 men and 108,771

women across Japan as a measure of HRV. Then, SDNN data were divided

into 47 prefectures, averaged over 4 seasons, and associated with the

prefectural averages of HALE reported by Japanese government. As the

results, in both sexes, prefectural age-adjusted HALE negatively correlated

with seasonal variation of prefectural age-adjusted SDNN, while it did not

corelated significantly with prefectural age-adjusted SDNN of any seasons.

Additionally, there was no significant correlation between prefectural

age-adjusted average life expectancy and the seasonal variation of

prefectural age-adjusted SDNN. Our observations support the hypothesis

that the magnitude of the seasonal variation in biological burden might be a

shortening factor for HALE.

K0063

Session 8

Presentation 3

(16:45-17:00)

Exploring Elders‘ Willingness and Needs for Adopting an Interactive

Somatosensory Game into Muscle Rehabilitation Systems

Yu-Ling Chu, Chien-Hsiang Chang, Yi-Wen Wang, Chien-Hsu Chen and

Yang-Cheng Lin

National Cheng Kung University, Taiwan

Abstract—Disease of the lower limb musculoskeletal system is one of the

most common diseases in elders. The use of interactive somatosensory

games (ISGs) in rehabilitation has been widely used. Most relevant studies

have focused on efficacy, while only a few have investigated the difficulties,

willingness, and requirements of elderly people in playing the game.

Therefore, this study is to design an ISG that is focused on lower-limb

rehabilitation and to explore whether the designed ISG can enhance the

willingness and motivation of elders to undergo rehabilitation. In this study,

15 elders (5 males and 10 females with average age of 78 years) with

degenerative joint disease were recruited to participate in our pretest-posttest

design experiments. First, the subjects completed a five-minute pre-test

questionnaire after a one-minute traditional rehabilitation (TR). Second,

they completed a five-minute post-test questionnaire after a one-minute

interactive somatosensory game rehabilitation (ISGR). The average

experiment time for each subject was 25 min. We used a t-test to analyze the

data. According to the result, there are significant differences in four factors

(i.e., interest (t=−6.89, p<0.05), self-affirmation (t=−3.17, p<0.05),

understanding of rehabilitation states (t=−3.31, p<0.05), and fatigue (t=2.49,

p<0.05)). The results show that (1) for the same treatment arrangement,

ISGR can make the elderly people more interested in and reduce their

fatigue while undertaking the long-term rehabilitation; (2) using ISGR can

increase their understanding of, and their confidence in, their rehabilitation

states; and (3) elders generally have positive attitudes toward the ISGR. The

results and analysis of willingness and needs in this study can be as a

reference for future studies of rehabilitation game development, and the

development of rehabilitation to improve the health of the elders.

Page 59: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 59 -

K2005

Session 8

Presentation 4

(17:00-17:15)

The Smartphone-based Rehabilitation Assistant for Stroke Patient

Pei-Yu Lin, Chieh-Min Lin, Chih-Liang Wu, Kuo-Feng Chou and Shih-Tsang

Tang

Ming Chuan University, Taiwan

Abstract—In addition to surgery and medicine, the rehabilitation is also an

important part for stroke treatment. However, it‘s a time-consuming and

tedious task. The physician usually ask patient for home rehabilitation.

Because the physician can not effectively realize patient‘s home condition,

the prognosis is poor. Therefore, this study developed a smartphone based

system for home rehabilitation assistant, which can record and store the

rehabilitation data, and then evaluate the outcome. This study used the

accelerometer module and MATLAB® to develop the prototype system,

which was for feasibility evaluation. And then the prototype system was

migrated to the smartphone. In the trial process, the subject was first to wear

the accelerometer module on the waist, and then walked in straight line and

went back. Next, the acceleration data were imported to MATLAB® for gait

analysis. After the prototype phase, the analysis algorithm was migrated to

the smartphone, and directly used the embedded accelerometer of

smartphone. In the aging society, the necessity of home rehabilitation is

increasing. Although this study is focusing on the stroke patient, but the

system structure and concept could also be a valuable reference in

developing the home rehabilitation system for other diseases.

K0060

Session 8

Presentation 5

(17:15-17:30)

Dynamically Colour Changing Actuator for Cyanosis Baby Manikin

Application with the Philips Hue LED Kit

Azmi Nur Fatihah, Frank Delbressine and Loe Feijs

Eindhoven University of Technology, Netherlands

Abstract—Medical simulation training is an important approach

contributing to medical safety. In the field of neonatology, serious efforts

have been devoted to construct manikins, which are realistic in shape and

improve the haptic experience. As central cyanosis is a severe pathological

sign in infants, new medical trainees or clinicians need to have

professional-level skills in assessing central cyanosis. In this paper, we are

trying out a new type of actuator, the Philips Hue LED bulb, to dynamically

generate a changing colour of cyanosis in a baby manikin. Philips Hue smart

lighting is widely used to remotely illuminate home and office. It is also

famous because of the wide range of white and coloured-LED bulbs, light

strips, and lamps which are used for changing the ambience and switching

on and off lights in response to the user input. In this paper, we describe a

dynamically colour changing actuator as the application of central cyanosis

in a newborn baby using Philip‘s Hue LED Kit. Herein, we demonstrated a

colourimetry analysis based on real cyanosis and non-cyanosis colour

changes on a 3D- printed baby‘s head. The colour change of the baby‘s head

was measured using the Spectraval 1501 spectroradiometer. In this study,

Page 60: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 60 -

the Hue lamp could be turned into a useful actuator to simulate cyanosis on

a baby. We measure the combined effects of the subtractive and additive

colour mixing which occur when adding an observation lamp. We believe

that these measurements will pave the way for a future implementation step,

in which the 3D printed head can be turned into a practical cyanosis

simulator to be used in real training.

K2007

Session 8

Presentation 6

(17:30-17:45)

A Deep Image based Proposal for Posture Classification

Wan-Chun Liao, Jing Wang, Yi-Shu Zhou, Ying-Hui Lai, Wai-Keung Lee

and Shih-Tsang Tang

Ming Chuan University, Taiwan

Abstract—Motion tracking is widely applied in military, entertainment,

sports, medical rehabilitation, computer vision, robotics, etc. The posture

classification is the basis technology for motion tracking. General posture

classification is implemented by wearable inertial device or RGB camera.

The inertial method need to be wore device in advanced, which is

inconvenient. The camera method is as well concerned in privacy. Our study

proposed an innovated method for posture classification. The ToF camera is

used to capture the deep image, which just acquires the distances data of the

reflected points on the object. As a result, there are no issues on the wearing

inconvenient and privacy. The artificial intelligent technology is then

applied for classification. In the trial, the subject is asked to pose before the

ToF camera. Then the deep image data are used to train the model for

posture classification. In the current status, we have successful discriminated

the postures of sitting, standing, and lying. Because of the proposed method

is convenient and private, which also shows a great potential in developing

the home rehabilitation assistant.

K0002

Session 8

Presentation 7

(17:45-18:00)

Rational Discovery of Human Superoxide Dismutase I (hSOD1)

Modulators: A Potential Therapeutic Target for Amyotrophic Lateral

Sclerosis (ALS)

Balasundaram Padmanabhan

National Institute of Mental Health and Neuro Sciences (NIMHANS), India

Abstract—Human Superoxide Dismutase I (hSOD1) is a cytoplasmic

homodimer enzyme involved in metabolizing superoxide radicals,

protecting cells from oxidative damage. The mutations in the SOD1 gene

have been linked to the neurodegenerative disease, familial amyotrophic

lateral sclerosis (fALS). The SOD1 toxic gain-of-function is caused by the

mutations in SOD1 which leads to get aggregation. Since both aggregation

propensity and protein stability strongly influence patient survival time after

onset of symptoms, working on protein stability is one of the better

strategies to improve the patient survival rate. The toxic-gain-of-function

can be prevented by stabilizing the hSOD1 dimer and/or by protecting the

Trp32 residue from oxidation. Hence, identifying and developing potential

library compounds to tackle the above-mentioned features are in need for

Page 61: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 61 -

the treatment of ALS. In this aspect, we have identified the ligands by

virtual screening using NCI-Diversity Set III and in-house library. The

SOD1-ligand complexes were crystallized, collected the X-ray diffraction

data on the beamline BM14 at ESRF, Grenoble, France. The structures of an

apo-form and protein-ligand complexes of hSOD1 refined to 1.9A

resolution and their binding studies would be discussed.

K0031

Session 8

Presentation 8

(18:00-18:15)

Orai3 Contributes to the Migration Capacity in Lobular Liver Cancer

Jing Yan, Xiuliang Zhao, Xia Liu, Zhaodi Gao and Ting Wei

Jining Medical University, China

Abstract—Store-operated calcium channels are known to be involved in

tumorigenesis, and one major component, calcium release-activated calcium

channel protein1 (Orai1), has already been extensively explored, whereas

the contribution of Orai3 to cancer remains unclear. The present study

investigated the role of Orai3 in lobular liver cancer. Western blot and

RT-PCR were utilized to assess Orai1/3 in patients and murine xenograft

models of liver cancer. Ca2+ imaging and whole-cell patch clamp were used

to examine store-operated Ca2+ entry (SOCE) activity. Transwell assays

and colony forming experiments were used to determine tumor

characteristics. We found that patients with lobular liver cancer displayed

high Orai1/3 levels, similar to that of murine xenograft models of lobular

liver cancer. Decreased Orai1/3 expression inhibited cancer volume in vivo

and suppressed colony formation and migration in vitro. However, only

Orai3-knockdown inhibited cancer metastasis in nude mice. Orai3

overexpression facilitated to the expression of E-cadherin in HepG2 and

SMMC7721 cells. Our findings suggest that liver cancer metastasis requires

Orai3 expression.

Page 62: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 62 -

Poster Session 1 Afternoon, January 20, 2020 (Monday)

Time: 15:00-16:00

Topic: “Biochemistry and Chemical Engineering”

Venue: Lobby of Room 3 (1F)

Session Chair: Prof. Ryoji Nagai

K0009

Poster 1

Mutation of Microbial Glycosyltransferase for Efficient and Regioselective

Biosynthesis of Rare Ginsenoside Rh1

Lu Zhao, Jianlin Chu and Bingfang He

Nanjing Tech University, China

Abstract—Ginsenoside Rh1 isolated from plant ginseng receives much

attention for its potential anticancer drug discovery. Enzymatic synthesis has

been developed as an alternative strategy to plant extraction and traditional

chemical synthesis of rare ginsenoside Rh1. Herein, we demonstrated the

glycosyltransferase UGTBL1 from Bacillus licheniformis which could

synthesize Rh1. Protopanaxatriol and UDP-glucose were docked into active

site of UGTBL1, and structure-based mutagenesis was carried out to explore

―hotspot‖ amino acids involved in substrate binding for regioselective

glycosylation of protopanaxatriol. Compared with wild type UGTBL1, the

mutants I62Y and I62H obtained by alanine scanning and saturation

mutations exhibited 12.89 and 13.31-fold regiospecificity towards

glycosylated product (Rh1), respectively. Moreover, mutants I62Y, I62H

also showed 1.51 and 1.44-fold conversion compared to wild type UGTBL1,

respectively. The exploratory mutational experiments of UGTBL1 indicated

that I62 is a key residue to efficient and regioselective glycosylation of

protopanaxatriol for ginsenoside Rh1 product. The mutants of UGTBL1

acquired in this study proved that the regioselectivity and glycosylation

activity of the enzyme for ginsenoside Rh1 synthesis could both be

enhanced. This may provide a new strategy for efficient biosynthesis of

ginsenoside Rh1.

K0061

Poster 2

The Biological Effects of Haematococcus Pluvialis Extracts on Fibroblasts

Lin Jun-Yan, Liu Wei-Chung adn Keng Nien-Tzu

Tzu Chi University of Science and Technology, Taiwan

Abstract—Aastaxanthin is a kind of antioxidant extracted from

haematococcus pluvialis. Research has indicated Aastaxanthin has the

function of anti-oxidation and would healing. The high level of reactive

oxygen species (ROS) will inhibit cell proliferation and reduce wound

healing; however, the bioeffects of Aastaxanthin on would healing are still

Page 63: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 63 -

unknown. This study focused on the ability of Aastaxanthin extracted from

Haematococcus pluvialis to promote would healing after cell damage. The

cytotoxicity of Astaxanthin extracts was evaluated by MTT assay. The safety

dose of extracts was added into the culture medium to investigate the effects

of ROS expression and cell migration rate in NIH 3T3 (mouse embryonic

fibroblast). The results showed that Astaxanthin extracts can increase cells

migration and proliferation. In addition, the ROS expression in NIH 3T3

cells was inhibited. The present results indicated that Astaxanthin extracts

can stimulate wound healing rate possibly via increasing the level of cells

migration and proliferation.

K0010

Poster 3

An Efficient Synthesis of Nucleoside Analogues Catalyzed by A Novel

Thermostable Purine Nucleoside Phosphorylase from Aneurinibacillus

Migulanus AM007

Gaofei Liu, Tiantong Cheng, Jianlin Chu, Sui Li and Bingfang He

Nanjing Tech University, China

Abstract—A novel purine nucleoside phosphorylase from Aneurinibacillus

migulanus AM007 (AmPNP) was cloned and heterologously expressed in

Escherichia coli BL21(DE3). The microbial AmPNP was identified as a

trimer and displayed low phosphorolysis activity toward adenosine. The

enzyme was thermostable at 55 °C for 12 h (retaining nearly 100% activity),

and efficiently biosynthesized adenosine analogs. A one-pot, two-enzyme

mode of trans-glycosylation was successfully constructed by combining

pyrimidine nucleoside phosphorylase derived from Brevibacillus

borstelensis LK01 (BbPyNP) and AmPNP for the production of purine

nucleoside analogs. The ratio of AmPNP and BbPyNP greatly influenced the

yield of purine nucleoside synthesis. The two-enzyme synthesis of

2-amino-6-chloropurine ribonucleoside catalyzed by BbPyNP and AmPNP

resulted in 95.1% conversion (20 mM sugar donor: 10 mM base acceptor;

1:1000 enzyme ratio). The conversions of 2,6-diaminopurine ribonucleoside

and 6-thioguanine ribonucleoside reached 98.3% and 92.3%, respectively.

The reaction of AmPNP coupled with BbPyNP could also efficiently

biosynthesize 2'-deoxyribonucleosides with the conversion of 96.5%

2,6-diaminopurine, 74.0% 2-amino-6-chloropurine, and 89.3%

6-thioguanine.

K3001

Poster 4

Repeated-batch Lactic Acid Fermentation using a Novel Bacterial

Immobilization Technique based on a Microtube Array Membrane

Chien-Chung Chen, Chuan-Chi Lan and Hong-Ting Victor Lin

National Taiwan Ocean University, Taiwan

Abstract—Lactic acid has received considerable attention for its various

applications. Although lactic acid is usually produced by batch fermentation

using free microbes, repeated-batch fermentation using immobilized cells

could offer several advantages over this method. In the present study, a

novel bacteria immobilization technique using a renewable poly-L-lactic

Page 64: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 64 -

acid (PLLA) microtube array membrane (MTAM) was thoroughly evaluated

in terms of its suitability for lactic acid fermentation. A bacteria

encapsulation efficiency of 85–90% was obtained using a siphon approach,

and bacteria in MTAMs with greater porosity showed greater lactic acid

productivity. MTAM-immobilized Lactobacillus acidophilus were

evaluated in eight cycles of repeated-batch lactic acid fermentation of

commercial medium MRS (4% glucose) and red seaweed Gracilaria

hydrolysate. The MTAM exhibited physical stability during lactic acid

fermentation and improved glucose consumption and lactic acid production

were achieved by immobilized cells. The immobilized bacteria showed a

maximum CLA of 37.39 g/L, r PLA of 0.79 g/L·h, and YL/S of 0.94 g/g in

MRS fermentation, and a maximum CLA of 27.76 g/L, r PLA of 0.58 g/L·h,

and YL/S of 0.92 g/g in seaweed hydrolysate fermentation. Our data indicate

that PLLA-MTAM is a novel, promising immobilization technology for

lactic acid fermentation.

K0011

Poster 5

A Simple Enhancer for Efficient Extracellular Expression of Human

Epidermal Growth Factor Secreted by E. Coli

Lupeng Cui, Yumeng Qiu and Bingfang He

Nanjing Tech University, China

Abstract—Human epidermal growth factor (hEGF) is a polypeptide (6.2

kDa) composed of 53 amino acids, and plays an important role to stimulate

the proliferation of various epidermal and epithelial tissues and is widely

used in clinical practices. However, the extracellular secretion strategy for

recombined hEGF expression by E. coli has rarely been successfully

achieved. In our previous study, we reported that a recombinant

β-fructofuranosidase (β-FFase) from Arthrobacter arilaitensis NJEM01 can

be efficiently secreted to culture medium with fused signal peptide by E.

coli. In present study, we exploited a simple enhancer which the truncated

N-terminal of signal peptide from β-FFase was fused to the N-terminal of

OmpA, and then the simple enhancer was further fused to the N-terminal of

hEGF to facilitate the expression of hEGF in E.coli. Interestingly, the

recombinant hEGF product was not only secreted to the periplasmic place

but also to the culture media by tricine-glycerin-SDS-PAGE analysis. The

biological activity of the extracellular expressed hEGF with simple

purification on cell proliferation was fully bioactive in vitro. These results

suggested that the simple enhancers introduced herein displayed the

secretary ability with signal peptide function and the hEGF protein was

successfully excreted to the culture medium by E. coli. As a result, the novel

enhancer was proved to be practically useful tool for the expressed

production of hEGF secretion.

K0065

Poster 6

Retinoic Acid Receptor - and Retinoid X Receptor -specific Agonists are

Hemodynamics-based Therapeutic Components for Vascular Disorders

Ting-Yu Lee

China University of Science and Technology, Taiwan

Page 65: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 65 -

Abstract—MicroRNA (miR)-

- inflammatory signaling in endothelial cells (ECs)

in response to atheroprotective pulsatile flow (PS). The aim of this study is

- -specific agonists can mimic the

effect of PS to induce miR-10a to repress inflammatory signaling and

vascular disorder. Our in vitro flow results showed that co-addition of

- -specific agonists has synergistic effect on rescuing

OS-inhibited EC miR-10a expression to repress inflammatory

GATA6/VCAM-1 signaling. Moreover, administration of ApoE-/-

mice with

- -specific agonists induced the expression of endothelial

miR-10a to repress GATA6/VCAM-1 signaling and inflammatory cell

infiltration in the vascular wall. In vivo inhibition of endothelial miR-10a by

- -specific

agonists. Our findings suggest that -specific agonists induce

EC miR-10a to inhibit the pro-inflammatory GATA6/VCAM-1 signaling

and inflammatory cell infiltration in the vessel wall and protect blood vessel

from vascular disorder.

K0044

Poster 7

The Mechanical Behaviors of Polyethylene/Silver Nanoparticle Composites:

An Insight from Molecular Dynamics Study

Chuan Chen and Shin-Pon Ju

Meiho University, Taiwan

Abstract—The mechanical properties of pristine polyethylene (PE) and its

composites with silver nanoparticles (PE/Ag NPs) at two Ag NP weight

fractions of 1.05 wt% and 3.10 wt% were studied by molecular dynamics

simulation (MD). It can be seen from the stress-strain distribution of the

tensile process that the embedded Ag NPs can significantly improve the

Young‘s modulus and tensile strength of the pristine PE, which is due to the

improve of the local density and strength of the PE near the Ag NP surface

in the range of 12 A� . Regarding the effect of temperature on the

mechanical properties of pristine PE and PE/Ag NP composites, the

Young‘s modulus and the strength of the pristine PE and PE/Ag NP

composites decreased significantly to 350 K and 450 K, consistent with

predicted melting temperature of pristine PE, which lies at around 360 K. At

such temperatures as these, PE material has stronger ductility and a higher

mobility of Ag NPs in the PE matrix than those at 300 K. With the increase

of tensile strain, Ag NPs tend to be close, and the fracture of PE leads to a

similarity between both the Young‘s modulus and ultimate strength found

for the pristine PE and those found for the PE/Ag NP composites at 350 K

and 450 K.

K2010

Poster 8

Characteristics of Hydrocarbons and Carbonyl Compounds Emitted from

the Oil Fume of Restaurant Oil Trap

Chia-Hsiang Lai, Yao-Kai Xiang and Zhen-Hong Zhu

Central Taiwan University of Science and Technology, Taiwan

Page 66: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 66 -

Abstract—The study investigated the emissions of the oil trap to

hydrocarbons, aldehydes and ketones at a university restaurant in central

Taiwan. Gaseous pollutions were collected over 15 working days in 2019.

The results found the highest levels of all restaurant oil fume was Methane

(19.352.43 ppm). The most abundant compounds of aldehydes were

2-Methylbutyaldehyde (7.920.09 ppm), followed by Formaldehyde,

Propionaldehyde and o-Tolualdehyde, with concentrations of 4.970.21,

2.293.22, and 2.110.13 ppm, respectively. The 9 compounds of ketones

were identified from the restaurant oil fume. The most abundant ketone

species were Methyl Ethyl Ketone (10.374.11 ppm) and Diethyl ketone

(2.081.32 ppm). The levels of the others ranged from 0.070.86ppm. The

total aldehydes concentration was approximately 2.6 times higher than that

for all ketones. The characteristics of hydrocarbons and carbonyl

compounds from restaurant oil fume might be affected by food, cooking

temperature and oil types. Comparison of emission levels of hydrocarbons,

aldehydes and ketones suggests that the most abundant contributions of

aldehydes from kitchen fume give priority to develop control devices.

K3004

Poster 9

Research on the Coordinated Development of Energy-environment System

Zhang Song

Research Institute of Petroluem Exploration& Development, China

Abstract—Energy-environment system is an open complex adaptive system

far from thermodynamic equilibrium. Based on the analysis of the features

of dissipative structure and key factors of entropy change in

energy-environment system, the coordinated development of

energy-environment system is studied in this paper. Furthermore, the

dissipative structure model is constructed to explore the operating and

evolutionary mechanism of energy-environment system. It also explores the

way to predict evolutionary trend of energy-environment system combined

with prediction theory. In the end, some strategic suggestions for developing

the energy-environment system are put forward. First, we should grasp the

evolution rule of energy-environment system. Second, we should construct

the innovation mechanism of energy-environment system and make full use

of the nonlinear function between elements to promote the formation of

self-organizing system. At last, it is important to further increase the

openness of the energy-environment system and promote the exchange of

material, energy, information and technology between internal and external

environment. The energy-environment system would accordingly become

more orderly and efficient.

K0018

Poster 10

Optimization of Cold-aDapted α-amylase Production in Escherichia Coli by

Regulation of Induction Conditions and Supplement with Osmolytes

Xiaofei Wang, Guangfeng Kan, Xiulian Ren, Cuijuan Shi and Ruiqi Wang

Harbin Institute of Technology at Weihai, China

Page 67: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 67 -

Abstract—In this study, the induction expression conditions of cold-adapted

α-amylase (Amy175) from Pseudoalteromonas sp. M175 in E. coli were

investigated. The optimal induction conditions were as follows:20 mM

L-proline was added into the culture medium at the initial inoculation time

point, the concentration of the inducer IPTG was 0.04 mM, the induction

point was 0.6 of OD600, the induction temperature was 15 °C, and the

induction time was 12 h, and the rotation speed was 120 rpm. The

maximum amylase activity could reach 141.1 U/ml at the optimal induction

conditions. Thus, the yield of the active and soluble α-amylase was found to

be increased in the lower induction temperature, lower IPTG concentration,

and the supplement of the proper amount of L-proline into the culture

medium.

K0037

Poster 11

Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung

Adenocarcinoma

Le Wei, Wanning Wen and Zhou Fang

The First Affiliated Hospital of Zhengzhou University, China

Abstract—Lung cancer is the most prevailing source of death due to cancer,

accounting for over 25% of death in the United States. Being able to predict

the survival time for patients will provide valuable information for the

choice of their treatment plans and benefit patient management. With the

advancement of next-generation sequencing, many high-throughput

sequencing data for DNA and RNA becomes available for cancer patients.

Here we present the results for using multiple machine learning algorithms

in predicting the survivorship of patients with Lung cancer adenocarcinoma.

Using the publicly available datasets in TCGA with the overall survival

length, and transcriptomic information, we evaluated our ability to predict

prognosis. We found that using the expression level of a few candidate

genes alone generates significant statistical power from a very limited

number of patients, suggesting more future studies to be conducted on

collecting such data to facilitate personalized medicine.

K1004

Poster 12

Identification of the Association between Hepatitis B Virus and Liver

Cancer using Machine Learning Approaches based on Amino Acid

Zhaoyang Cao

University of British Columbia, Canada

Abstract—Primary liver cancer has been a common reason for death from

cancer globally. The most common type of primary liver cancer is the

hepatocellular carcinoma (HCC). The major cause of HCC is chronic

infections with hepatitis B virus (HBV). In this research, we used next

generation sequencing (NGS), which has been very widely used to produce

deep, efficient, and high-quality sequence data. NGS was used to sequence

the pre-S region of the HBV genome of total 139 patients, which contain 94

HCC patients and 45 chronic HBV (CHB) patients. We generated two types

of datasets. Firstly, for the data of amino acid occurrence frequency, we used

Page 68: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 68 -

basic local alignment search tool (BLAST) to map each NGS short read and

translated each alignment into amino acid by DNA codon table. The input

features are the occurrence frequencies of 20 basic amino acids using

Shannon entropy. We picked 40 patients with 27 HCC and 13 CHB as the

independent testing set. Then we used machine learning methods including

logistic regression, random forest and support vector machine (SVM) to

construct the classification models and make the prediction. The AUC

values on the independent testing set for those machine learning methods

(logistic regression, random forest and SVM) are 0.946, 0.923 and 0.960

respectively. Secondly, for the data of word pattern frequency of amino

acids, we calculated word pattern frequencies of amino acids of all

individuals and compared them using Euclidean distance. The input features

are the frequencies of amino acid word of length 2, which is normalized by

dividing the total occurrence number of all words. What‘s more, word

pattern frequencies of amino acids were used to construct the classification

models for HCC status using machine learning methods. Principal

coordinate analysis (PCoA) was also used to visualize the associations

between patient clusters, the HCC disease status of patients, and the fraction

of HBV genotypes. We found that word patterns are powerful for the

analysis of the HBV sequences from the aspect of amino acids because the

AUC values of the classification models for machine learning methods are

all above 0.9. Hence, our study showed that word pattern frequencies of

amino acids is powerful for revealing the underlying principles of the

occurrence of HCC triggered by HBV. Our essential findings consist of three

parts. Firstly, all machine learning methods can generate classification

models with high AUC values. Then, we can find some certain positions of

amino acids or word patterns of amino acids that the mutation occurred on

those positions will induce the HCC. Last, PcoA is associated with the

disease status (HCC or CHB) and the fraction of genotype B (or C).

K0070

Poster 13

Computational Modeling of Myocardial Thermal Lesion Induced by

Multi-source Frequency Control RF ablation Method

Shengjie Yan, Kaihao Gu, Weiqi Wang and Xiaomei Wu

Fudan University, China

Abstract—In minimally invasive surgery for atrial fibrillation,

radiofrequency (RF) voltage is usually used to ablate cardiac tissue. In this

study, a computational modeling of multi-source frequency control RF

ablation mode (FcM) was constructed to analyze the characteristics of

thermal field and electric field in myocardium. To highlight the relationship

between frequency difference and lesion size of ablation, the amplitude and

phase angle of voltage in the simulation were fixed at 20 V peak and 0

degree, respectively. Furthermore, the thermal lesion depth, width and

continuity produced by FcM and multi-source bipolar ablation mode(BiM)

were also compared in this paper. At the electrical boundary condition for

creating continuous lesion, the transmural and continuous lesion (long or

Page 69: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 69 -

short) was generated by FcM, whereas BiM did not create transmural lesion.

At the electrical boundary condition for creating discrete lesion, FcM was

able to create two or one transmural discrete lesion, whereas BiM did not

create transmural lesion. Simulation results show that a kind of transmural,

continuous, symmetrical lesion shape was generated in atrial tissue using

multi-source frequency control RF ablation and the lesion size was larger

than BiM at identical ablation condition.

Page 70: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 70 -

Poster Session 2 Afternoon, January 21, 2020 (Tuesday)

Time: 15:15-16:15

Topic: “Systematic Biology”

Venue: Lobby of Room 3 (1F)

Session Chair: Prof. Masako Takasu

K0083

Poster 1

CitAB Two-component System-regulated Citrate Utilization Contributes to

Vibrio Cholerae Competitiveness with the Gut Microbiota

Ming Liu

NanJing Agricultural University, China

Abstract—Citrate is a ubiquitous compound and can be utilized by many

bacterial species, including enteric pathogens, as a carbon and energy

source. Genes involved in citrate utilization have been extensively studied in

some enteric bacteria, such as Klebsiella pneumoniae, however, their role in

pathogenesis is still not clear. In this study, we investigated citrate utilization

and regulation in Vibrio cholerae, the causative agent of cholera. The

putative anaerobic citrate fermentation genes in V. cholerae, consisting of

citCDEFXG, citS-oadGAB, and two-component system (TCS) genes citAB,

are highly homologous to those in K. pneumoniae. Deletion analysis shows

that these cit genes are essential for V. cholerae growth when citrate is the

sole carbon source. The expression of citC and citS operons was dependent

on citrate and CitAB, whose transcription was autorepressed and regulated

by another TCS regulator ArcA. In addition, citrate fermentation was under

the control of catabolite repression. Mouse colonization experiments showed

that V. cholerae can utilize citrate in vivo using the citrate fermentation

pathway and that V. cholerae likely needs to compete with other members of

the gut microbiota to access citrate in the gut.

K2002

Poster 2

Preparation and Application of Chitosan-phytic Acid Complex Gel Beads

Shao-Jung Wu, Zhi-Run Chen and Chi-Chen Kuo

Ming Chi University of Technology, Taiwan

Abstract—A novel and straightforward synthetic technique for the

preparation of chitosan beads with less cytotoxicity is desirable. Chitosan

can form gels with the nontoxic and multivalent inositol hexakisphosphate

counter ion. The polycationic chitosan and polyanionic phytic acid via an

ionotropic gelation process to form gels. Crosslinking characteristics of the

chitosan-polyanions beads were improved by the modification of in-liquid

curing mechanism of the beads. The present study aims to explore the

adsorption capacities and kinetics of Cu(II) ions from aqueous solution onto

Page 71: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 71 -

fabricated gel beads. The influence of initial pH of the metal ions solution

and initial concentration of metal ions solution on the uptake of metal ions

were also studied. The equilibrium adsorption data of Cu(II) ions on the gel

beads were correlated well with Langmuir isotherm model with a maximum

adsorption capacity of 188.68 mg/g. The adsorption of Cu(II) ions on the gel

beads exhibited pseudo-second order kinetics. The adsorption

thermodynamic parameters indicated that the involved process should be

spontaneous and endothermic. This study includes a comparative study on

the adsorption capacity of chitosan-based adsorbent, and the results for the

application of chitosan-polyanions beads are evaluated in future wastewater

treatments.

K0084

Poster3

The 58th Cysteine of TcpP is Essential for Vibrio Cholerae Virulence Factor

Production and Pathogenesis

Mengting Shi

NanJing agricultural university, China

Abstract—Vibrio cholerae, the causative agent of the severe diarrheal

disease cholera, has evolved signal transduction systems to control the

expression of virulence determinants. It was previously shown that two

cysteine residues in the periplasmic domain of TcpP are important for TcpP

dimerization and activation of virulence gene expression by responding to

environmental signals in the small intestine such as bile salts. In this study,

the functions of cysteine residues in TcpP cytoplasmic domain were

investigated and we found that only C58 is essential for TcpP dimerization

and for activating virulence gene expression. To better characterize this

cysteine residue, site-directed mutagenesis was performed to assess the

effects on TcpP homodimerization and virulence gene activation. A

TcpPC58S mutant was unable to form homodimers and activate virulence

gene expression, and did not colonize infant mice. However, a

TcpPC19/51/124S mutant was not attenuated for virulence. These results

suggest that C58 of TcpP is indispensable for TcpP function and is essential

for Vibrio cholerae virulence factor production and pathogenesis.

K0053

Poster 4

Laboratory Solution Service using ExTOPE® IoT Portal

Kazuyuki Kato, Zentaro Kadota, Shoichi Uemura, Masataka Okayama,

Hiroshi Yamaguchi and Tomohiro Araki

Hitachi High-Tech Fielding Corporation, Japan

Abstract—We are developing a protein identification system by amino acid

composition analysis. This system identifies a protein automatically by

comparing the amino acid analysis results with Uni-Prot array data. This

protein identification method is unique as using the total amino acid

composition information not the amino acid sequence information such as

mass analysis or classical protein sequencer analysis. For the data

processing using such a protein identification system, we focused on

improving the data treatments. To improve the data accessibility, ExTOPE

Page 72: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 72 -

system enables to worldwide data processing and data sharing. Namely,

ExTOPE® served by cloud computing has 3 useful functions; automatic

data gathering by measurement the results and error logs, operating by

remote control, and connecting with other applications by application

interfaces. Therefore, ExTOPE® could be a component for automatic

system which identified proteins after upload amino acid composition

analysis data into cloud computing. Besides, we could use remote control

and increase stable quality of machines. As a result, we suggest serviceable

laboratory solution to increase useful value of amino acid analyzers with

ExTOPE®. We propose ExTOPE® as a laboratory solution system that

maximize the utility value of the equipment.

K1012

Poster 5

Procleave: A Bioinformatic Approach for Protease-specific Substrate

Cleavage Site Prediction by Combining Sequence and Structural

Information

Fuyi Li, Tatsuya Akutsu, Jian Li and Jiangning Song

Monash University, Australia

Abstract—Proteases are enzymes that cleave and hydrolyse the peptide

bonds between two specific amino acids of target substrate proteins.

Protease-controlled proteolysis plays a key role in the degradation and

recycling of proteins, which is essential for various physiological processes.

Thus, solving the substrate identification problem will have important

implications for the precise understanding of protease functions and their

physiological roles, as well as for therapeutic target identification and

pharmaceutical applicability. Consequently, there is a great demand for

bioinformatics methods that can predict novel substrate cleavage events with

high accuracy by utilizing both sequence and structural information. In this

study, we present Procleave, a novel bioinformatics approach for predicting

protease-specific substrates and specific cleavage sites by taking into

account both sequence and 3D structural information. Structural features of

known cleavage sites were represented by discrete values using a LOWESS

data-smoothing optimization method, which turned out to be critical for the

performance of Procleave. The optimal approximations of all structural

parameter values were encoded in a conditional random field (CRF)

computational framework, alongside sequence and chemical-group-based

features. Here, we demonstrate the outstanding performance of Procleave

through extensive benchmarking and independent tests. Procleave is capable

of correctly identifying most cleavage sites in case study proteins.

Importantly, when applied to the human structural proteome encompassing

17,628 protein structures, Procleave suggests a number of potential novel

target substrates and their corresponding cleavage sites of different

proteases. Procleave has been implemented as a webserver and is freely

accessible at http://procleave.erc.monash.edu/.

K1005 BnMTP3, An Intracellular Transporter from Brassica Napus Confer Zn and

Mn Tolerance in Arabidopsis Thaliana

Page 73: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 73 -

Poster 6 Dongfang Gu, Xueli Zhou, Xi Chen and Wei Zhang

Nanjing Agricultural University, China

Abstract—The micronutrient elements Zn and Mn are required for plant

growth and development. As an important oil crop in the world, the yield

and quality of rapeseed are affected by Zn or Mn toxicity. The CDF family

plays an important role in maintaining intracellular ionic homeostasis and

tolerance in model plants, but it has rarely been reported in rapeseed. In this

study, the functional characterisation of a Brassica napus cation diffusion

facilitator (CDF/MTP) protein BnMTP3, homologous to AtMTP3, is

investigated. The heterologous expression of BnMTP3 in yeast enhances

tolerance and intracellular sequestration to Zn and Mn. The

MTP3-dependent increase in metal tolerance was more pronounced for Zn

than for Mn. Expression of BnMTP3 increased Zn and Mn tolerance in

Arabidopsis thaliana, but also markedly increased Zn accumulation in roots.

This is consistent with the results of expression analysis that BnMTP3 was

primarily expressed in roots. Subcellular localisation suggested that

BnMTP3 was localised to TGN and PVC, and fluorescent signals were

observed on vacuolar membranes after treatment with Zn and Mn. These

findings indicated that BnMTP3 is a member of Zn-CDFs that may

sequester excess Zn and/or Mn into vacuoles.

K0055

Poster7

Comparing Dissimilarity Metrics for Clustering Gene into Functional

Modules using Machine Learning

Xin Yan and Dantong Lyu

Sun Yat-Sen University, China

Abstract—Clustering is widely used in biological analyses for clustering

genes into functional modules. For any clustering mechanism, we need to

define some measurements for dissimilarity. The two most commonly used

dissimilarity metrics are the Manhattan distance and Euclidean distance.

Moreover, the 1-correlation coefficient is also commonly used for defining

similarity. Here, we use the transcriptomic data across multiple

environments in yeast for gene clustering and evaluate the performance of

using these four dissimilarity metrics. We designed two metrics that use

1-abs(correlation) for Pearson and Spearman correlation. We found that

1-abs(Pearson correlation) works the best in two test cases for identifying

genes involved in ethanol metabolism and galactose metabolism and build a

clustering model based on the metric. We propose that this dissimilarity

metric be used for future studies of clustering of genes based on expression

level. Such information, combined with more gathering of transcriptomic

information across environments, will boost our understanding of gene

clustering and modularity in exploring unknown species.

K1006

Poster 8

OsGCT1, A Novel Gene Related to Cadmium, Copper Tolerance and

Accumulation in Rice

Ending Xu, Xi Chen and Wei Zhang

Page 74: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 74 -

Nanjing Agricultural University, China

Abstract—Copper (Cu) is an essential micronutrient for plants and humans.

Cadmium (Cd) is a Cu analog and one of the most toxic heavy metals to

humans. Here we investigated the role of the Cu/Cd transporter OsGCT1.

OsGCT1:GFP fusion protein localized to the Golgi in rice protoplast. The

growth of yeast expressing OsGCT1 was impaired in the presence of Cd and

Cu compared with yeast transformed with an empty vector. Moreover, the

Cd and Cu content of OsGCT1-expressing yeast exceeded that of the vector

control. The expression of OsGCT1 in rice was observed mainly in the

leaves where OsGCT1 transcripts were abundant in vascular bundles.

Knockdown of OsGCT1 resulted in growth inhibition in the presence of

high concentrations of Cd and Cu, and also led to increased accumulation of

Cd and Cu in the shoots and new leaves. The overexpression of OsGCT1

was found to enhance Cd and Cu tolerance in rice. These findings suggest

that OsGCT1 is a Golgi-localized Cd and Cu transporter that is required for

Cd and Cu homeostasis and contributes towards Cd and Cu tolerance in rice.

K1001

Poster 9

Rectal Methods of Delivery of Medical Drugs of the Protein Nature

J.K. Ukibayev, U.M. Datkhayev, A.P. Frantsev and D.A. Myrzakozha

Kazakh National Medical University named after S.D. Asfendiyarov,

Kazakhstan

Abstract—Intravenous injection of protein drugs causes many negative side

effects known as infusion reactions and serious consequences, such as serum

sickness. This article shows the possibility of intake into the body of

recipients of drug preparations of protein nature in the form of rectal

suppositories. The use of rectal forms of protein preparations has a number

of significant advantages compared with injectable forms. They are more

effective than injectable forms, since their active molecules are less exposed

to the destructive action of liver enzymes. Rectal forms of protein

preparations are easy to use for children and elderly patients, they are

absolutely safe for nosocomial infections - syphilis, hepatitis, HIV, etc.

Moreover, they are simpler in the technology of their manufacture.

K2018

Poster 10

The Application of Designated Manager System on Protection Forest

Management at Saga Prefecture, Japan - from the Public-Private Partnership

and Policy Implementation Approaches

Ching Li, Chiachi Cheng and Yi-Chine Chu

Industrial Technology Research Institute, Taiwan

Abstract—In 2003, Japanese government pointed out the Deigned Manager

System to emphasize the importance of combining national and private

ability to raise the public benefits and increase citizen participation rates. In

addition, Forest Agency hopes to make use private ability to realize the

diversified management of forest business, those are the backgrounds of

public-private partnerships of forest management in Japan. The purpose of

Page 75: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 75 -

this study was to discuss the application of the Designated Manager System

on protection forest management at Saga Prefecture, Japan. The study took

Nijinomatsubara Pine Forest and Saga 21 Century Kenminnomori Forest

Learning Center as the examples, attempted to define the roles of each

organization, the public values, and the co-governors on the policy

implication processes. This study applied document analysis and interview

with government officers and NPO managers. This study found, the

governments played different roles to integrate process for different public

value, in Nijinomatsubara, local government plays the role as a

communication person between government and locals; in the case of

Kenminnomori Forest Learning Center, local government plays the role as

a provider, provides the wellness place to NPO and citizens for learning

forest related knowledge. Also, vary with issue initiation, each NPO niche

abilities for co-governers, Nijinomatsubara‘s NPO have basically ecological

environment knowledge and activity integration ability, the NPO of Saga

Forest Learning Center have professional forest related knowledge and

woodworking ability. Both of these 2 cases aim to widen range of public

participation. Therefore, PPP not only the idea for taking advantage of

private organization or the NPO only want to get benefits from

governments, but share the value within government, NPO and citizens, and

widen the range of public participation.

K0067

Poster 11

Automatic Brain Mask Segmentation for Mono-modal MRI

Yanwu Yang, Chenfei Ye, Xutao Guo, Chushu Yang and Heather T. Ma

Harbin Institute of Technology, Shenzhen, China

Abstract—In recent years, deep learning methods have gained promising

results in different kinds of image processing tasks, such as image

classification, semantic segmentation, image generation and so on. This

paper focuses on the research of brain masking for mono-modal MRI,

structural MRI, which is the most commonly used by the clinic and

research. The brain mask is a basic and essential tool for brain function

analysis and voxel-based structural analysis. In this paper, we present an

automatic method for brain masking which would match the brain atlas for

the origin image and also extract the regions of interest (ROI), like

Hippocampus. Our network is developed from the U-net and a coarse mask

is added into the network, which is generated by the method of region seeds

growing. The combination of coarse mask and origin input speeds up the

localization of the network and also increases the segmentation accuracy. In

this work, two groups of experiments have been carried out, the one to do

the brain mask automatically for the whole brain and the other for the region

of Hippocampus extraction. Finally we have gained 0.893 dice coefficient

for Hippocampus and 0.865 for the whole brain regions in average

K0056

Poster 12

The Application and Comparison of Confocal and SIM Imaging System

Yiran Liu

University of California, USA

Page 76: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 76 -

Abstract—Optical imaging is a popular method for biology research, and

now there lots of optical imaging system. Confocal imaging is a wide used

one which was introduced the pinhole and subsequently is not affected by

the out-of-focus signal, making the method be better than widefield

microscopy. It is acceptable that confocal microscopy has its own limits in

resolution. Researchers struggled to push the resolution to another limit

during the process many fancy optical imaging methods were invented, such

as structured illumination microscopy (SIM). We are here to apply the two

optical imaging methods in RyR2 labelled heart cells to compare the two

methods, and we found that confocal imaging does show the disadvantages

such as photobleaching and limited resolution, while SIM imaging has a

higher resolution to observe much more details of the RyR2 location.

K0082

Poster 13

Colon Cancer Inhibition Capacity Evaluation and Mechanism Study of A

Novel Biopolymer Conjugated Gold Nanoparticle System as Chemo-reagent

Carrier

Wei-Hung Hung, Kuen-Chan Lee and Er-Chieh Cho

Taipei Medical University, Taiwan

Abstract— Colon cancer is a common leading cause of death around the

world, especially in western countries. Even though great progress has been

made for colon cancer therapeutic development, there are still clinical

challenges in treating late staged or metastatic patients. Nanomaterials have

been intensively applied in biomedical studies. Gold nanoparticles have

been demonstrated as a promising drug carrier with great cellular

biocompatibility, yet little studies examined the capacity of biopolymer

conjugated gold nanoparticle in colon cancer. In this study, we assembled

and investigated the system with chemo-reagent in both cancer cell model

and mice model. The system was evaluated by UV spectrum and

transmission electron microscope, etc. first, and then the cancer suppression

capacity of the system and regulatory mechanisms have been studied by

different molecular biological assays. In conclusion, our results showed that

our novel drug carrier system is powerful in treating colon cancer, and that

this system can potentially contribute to other diseases in the future.

Page 77: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 77 -

Academic Visit

9:30-17:30, January 22, 2020 (Wednesday)

Tip: 1. Gather at 京都駅八条口 日本〒601-8003 Kyoto, Minami Ward, Higashikujo Nishisannocho. (In

front of 7-ELEVEN Convenience Store.) before 9:30 a.m.

2. The quotation only includes lunch, admission tickets of Kinkaku-ji Temple and Kiyomizu Temple.

3. The following places are for references, and the final schedule should be adjusted to the actual notice.

Time Specific Arrangement

9:30-10:00 Depart at 京都駅八条口

10:30-12:00 Enjoy the breathtaking Kinkaku-ji Temple - the golden pavilion

12:00-15:30

Lunch

Discover the amazing Kiyomizu Temple

Appreciate handicrafts at Ninen-zaka and Sannen-zaka

Photograph at Ishibe Alley

Stroll around Gion (The Geiko/Geisha District)

Visit Yasaka Shrine

15:30-17:00 Walk among vermillion gates at Fushimi Inari

17:00-17:30 Return

Kinkakuji is a Zen temple in northern Kyoto whose top two

floors are completely covered in gold leaf. Kinkakuji is an

impressive structure built overlooking a large pond, and is

the only building left of Yoshimitsu's former retirement

complex. It has burned down numerous times throughout its

history including twice during the Onin War, a civil war that

destroyed much of Kyoto; and once again more recently in 1950 when it was set on fire by a

fanatic monk. The present structure was rebuilt in 1955.

Kiyomizu (literally "Pure Water Temple") is one of the most

celebrated temples of Japan. It was founded in 780 on the

site of the Otowa Waterfall in the wooded hills east of Kyoto,

and derives its name from the fall's pure waters. The temple

was originally associated with the Hosso sect, one of the

oldest schools within Japanese Buddhism, but formed its

own Kita Hosso sect in 1965. In 1994, the temple was added to the list of UNESCO world

heritage sites.

Page 78: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 78 -

Ninen-zaka and Sannen-zaka Two of Kyoto‘s most attractive

streets are Sannen-zaka and Ninen-zaka, a pair of gently sloping

lanes that lead down from Kiyomizu-dera Temple toward

Nene-no-Michi Lane. Lined with beautifully restored traditional

shophouses and blissfully free of the overhead power lines that

mar the rest of Kyoto, Sannen-zaka and Ninen-zaka are a pair of

pedestrian-only lanes that make for some of the most atmospheric strolling in the whole city.

Ishibe Alley is a must-see in Kyoto. As you walk through the alleyways you feel like you've

stepped into a movie set, only it's real, because the establishments in Ishibe are not just for

display.

Gion is Kyoto's most famous geisha district, located around

Shijo Avenue between Yasaka Shrine in the east and the Kamo

River in the west. It is filled with shops, restaurants and ochaya

(teahouses), where geiko (Kyoto dialect for geisha) and maiko

(geiko apprentices) entertain.Gion attracts tourists with its high

concentration of traditional wooden machiya merchant houses.

Yasaka Shrine, also known as Gion Shrine, is one of the most famous shrines in Kyoto.

Founded over 1350 years ago, the shrine is located between

the popular Gion District and Higashiyama District, and is

often visited by tourists walking between the two

districts.The shrine's main hall combines the honden (inner

sanctuary) and haiden (offering hall) into a single building.

In front of it stands a dance stage with hundreds of lanterns

that get lit in the evenings. Each lantern bears the name of a local business in return for a

donation.

Fushimi Inari Shrine is an important Shinto shrine in

southern Kyoto. It is famous for its thousands of vermilion

torii gates, which straddle a network of trails behind its

main buildings. The trails lead into the wooded forest of the

sacred Mount Inari, which stands at 233 meters and belongs

to the shrine grounds. Fushimi Inari is the most important of

several thousands of shrines dedicated to Inari, the Shinto god of rice. Foxes are thought to be

Inari's messengers, resulting in many fox statues across the shrine grounds. Fushimi Inari

Shrine has ancient origins, predating the capital's move to Kyoto in 794.

Page 79: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 79 -

Note

Page 80: CONFERENCE ABSTRACT - ICEBE · K0037: Using Multiple Machine Learning Algorithms for Cancer Prognosis in Lung Adenocarcinoma Le Wei, Wanning Wen and Zhou Fang 67 K1004: Identification

ICEBE 2020 CONFERENCE ABSTRACT

- 80 -

Note